{"id":"56693b42-d0d3-4bfb-a37d-7e2cf83597e1","entityType":"agent","slug":"clawhub-dlazyai-dlazy-image-social-media","name":"社交媒体配图设计 Social Media Design","canonicalUrl":"https://www.xpersona.co/agent/clawhub-dlazyai-dlazy-image-social-media","canonicalPath":"/agent/clawhub-dlazyai-dlazy-image-social-media","generatedAt":"2026-10-10T10:46:10.192Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T04:52:37.536Z","emptyReason":null},"description":"A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs...","descriptionLabel":"Source description","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 4.8K downloads reported by the source. Last updated 10/9/2026.","installCommand":"clawhub skill install s170j1ymymrxasgd00dsk7tckx84cf45:dlazy-image-social-media","sourceUrl":"https://clawhub.ai/dlazyai/dlazy-image-social-media","homepage":"https://clawhub.ai/dlazyai/skills/dlazy-image-social-media","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/dlazyai/dlazy-image-social-media","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/dlazyai/skills/dlazy-image-social-media","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":74,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"社交媒体配图设计 Social Media Design technical dossier on Xpersona with agent coverage, OPENCLEW support, and live trust metadata."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-09T04:52:37.536Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T04:52:37.536Z","emptyReason":null},"stars":null,"forks":null,"downloads":4773,"packageName":null,"latestVersion":"1.3.27","tractionLabel":"4.8K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T04:52:37.535Z","emptyReason":null},"lastUpdatedAt":"2026-10-09T04:52:37.536Z","lastCrawledAt":"2026-10-09T04:52:37.535Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-10T04:52:37.535Z","lastVerifiedAt":null,"highlights":[{"version":"1.3.27","createdAt":"2026-10-08T01:15:15.591Z","changelog":"例行版本更新 2026-10-08","fileCount":4,"zipByteSize":12016},{"version":"1.3.26","createdAt":"2026-10-04T01:14:04.804Z","changelog":"例行版本更新 2026-10-04","fileCount":4,"zipByteSize":12041},{"version":"1.3.25","createdAt":"2026-10-02T04:56:34.823Z","changelog":"例行版本更新 2026-10-02","fileCount":4,"zipByteSize":12029},{"version":"1.3.24","createdAt":"2026-09-30T01:12:57.054Z","changelog":"例行版本更新 2026-09-30","fileCount":4,"zipByteSize":11998},{"version":"1.3.23","createdAt":"2026-09-28T02:11:56.162Z","changelog":"例行版本更新 2026-09-28","fileCount":4,"zipByteSize":11879},{"version":"1.3.22","createdAt":"2026-09-28T01:13:55.461Z","changelog":"例行版本更新 2026-09-28","fileCount":4,"zipByteSize":12109},{"version":"1.3.21","createdAt":"2026-09-24T06:49:49.982Z","changelog":"例行版本更新 2026-09-24","fileCount":4,"zipByteSize":12183},{"version":"1.3.20","createdAt":"2026-09-22T01:11:37.935Z","changelog":"例行版本更新 2026-09-22","fileCount":4,"zipByteSize":12213}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s170j1ymymrxasgd00dsk7tckx84cf45:dlazy-image-social-media","setupComplexity":"low","setupSteps":["Install using `clawhub skill install s170j1ymymrxasgd00dsk7tckx84cf45:dlazy-image-social-media` in an isolated environment before connecting it to live workloads.","No published capability contract is available yet, so validate auth and request/response behavior manually.","Review the upstream CLAWHUB listing at https://clawhub.ai/dlazyai/dlazy-image-social-media before using production credentials."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T10:46:10.188Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-dlazyai-dlazy-image-social-media/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T04:52:37.536Z","emptyReason":null},"readme":"Skill: 社交媒体配图设计 Social Media Design\n\nOwner: dlazyai\n\nSummary: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs...\n\nTags: latest:1.3.27\n\nVersion history:\n\nv1.3.27 | 2026-10-08T01:15:15.591Z | user\n\n例行版本更新 2026-10-08\n\nv1.3.26 | 2026-10-04T01:14:04.804Z | user\n\n例行版本更新 2026-10-04\n\nv1.3.25 | 2026-10-02T04:56:34.823Z | user\n\n例行版本更新 2026-10-02\n\nv1.3.24 | 2026-09-30T01:12:57.054Z | user\n\n例行版本更新 2026-09-30\n\nv1.3.23 | 2026-09-28T02:11:56.162Z | user\n\n例行版本更新 2026-09-28\n\nv1.3.22 | 2026-09-28T01:13:55.461Z | user\n\n例行版本更新 2026-09-28\n\nv1.3.21 | 2026-09-24T06:49:49.982Z | user\n\n例行版本更新 2026-09-24\n\nv1.3.20 | 2026-09-22T01:11:37.935Z | user\n\n例行版本更新 2026-09-22\n\nv1.3.19 | 2026-09-20T01:17:23.472Z | user\n\n例行版本更新 2026-09-20\n\nv1.3.18 | 2026-09-14T01:13:56.287Z | user\n\n例行版本更新 2026-09-14\n\nv1.3.17 | 2026-09-10T01:11:09.992Z | user\n\n例行版本更新 2026-09-10\n\nv1.3.16 | 2026-09-08T01:11:52.888Z | user\n\n例行版本更新 2026-09-08\n\nv1.3.15 | 2026-09-07T01:15:22.830Z | user\n\n例行版本更新 2026-09-07\n\nv1.3.14 | 2026-09-04T01:17:22.997Z | user\n\n例行版本更新 2026-09-04\n\nv1.3.13 | 2026-09-02T01:11:53.395Z | user\n\n例行版本更新 2026-09-02\n\nv1.3.12 | 2026-08-31T02:52:31.747Z | user\n\n源码仓库迁移至 github.com/dlazy-ai/cli\n\nv1.3.11 | 2026-08-24T09:23:27.112Z | user\n\nRestore Chinese display name\n\nv1.3.10 | 2026-08-17T09:06:01.073Z | user\n\nRestore Chinese display name\n\nv1.3.9 | 2026-08-10T05:47:14.248Z | user\n\nRestore Chinese display name\n\nv1.3.8 | 2026-08-10T04:01:46.389Z | user\n\nRestore Chinese display name\n\nv1.3.7 | 2026-07-21T07:54:12.267Z | auto\n\n- Updated documentation in SKILL.md and SKILL-cn.md.\n- Removed obsolete skill-card.md file.\n- No functional changes; documentation only.\n\nv1.3.6 | 2026-07-21T02:18:26.748Z | auto\n\n- Updated CLI dependency version from 1.2.2 to 1.2.3 throughout documentation and install instructions.\n- Removed outdated file: skill-card.md.\n- SKILL.md and SKILL-cn.md updated to reflect new CLI version and installation guidance.\n- No changes to workflow or execution steps.\n\nv1.3.5 | 2026-07-20T05:53:15.301Z | user\n\n规范 dlazy- 前缀命名，补充中文显示名\n\nv1.3.4 | 2026-07-17T10:38:27.615Z | auto\n\n- Raised version to 1.3.4.\n- Updated documentation in SKILL.md and SKILL-cn.md.\n- Removed obsolete file: skill-card.md.\n\nv1.2.2 | 2026-07-17T08:44:25.820Z | auto\n\n- Dropped the file skill-card.md.\n- Updated documentation in SKILL.md and SKILL-cn.md.\n- Rolled back the skill version in SKILL.md from 1.2.2 to 1.2.1.\n- No functional or user-facing logic changes.\n\nv1.3.3 | 2026-07-09T01:48:11.368Z | user\n\nset Chinese display name for CN search\n\nv1.3.2 | 2026-07-09T01:47:59.418Z | user\n\nset Chinese display name for CN search\n\nv1.3.1 | 2026-07-09T01:30:21.296Z | user\n\nset Chinese display name for CN search\n\nv1.3.0 | 2026-07-07T19:44:30.093Z | auto\n\ndlazy-image-social-media v1.3.0\n\n- Updated documentation files (SKILL.md and SKILL-cn.md) to clarify feature descriptions and skill usage.\n- No changes to functionality; updates are documentation-only.\n\nv1.2.1 | 2026-07-07T18:32:45.807Z | auto\n\nVersion 1.2.1 of dlazy-image-social-media\n\n- Updated skill metadata and name to include more platforms and bilingual description.\n- SKILL.md and SKILL-cn.md updated for improved clarity and completeness.\n- Removed obsolete skill-card.md file.\n- Minor edits to platform list and wording for better transparency and accuracy.\n\nv1.2.0 | 2026-06-02T10:04:26.557Z | auto\n\n- CLI dependency version updated from `@dlazy/cli@1.0.9` to `@dlazy/cli@latest` for global install and usage instructions.\n- Metadata fields and CLI install instructions changed to use the latest CLI version.\n- Documentation now consistently refers to the latest CLI version in code snippets and install guidance.\n- Obsolete file `skill-card.md` removed from the skill package.\n\nv1.1.2 | 2026-06-02T01:40:44.446Z | user\n\nUpdate model skills (names/params refreshed); add search-audio/image/video\n\nv1.1.1 | 2026-05-06T14:13:51.100Z | auto\n\n- Documentation updated; no code or functional changes.\n- Minor version bump to 1.1.1 for clarity.\n- No user-facing changes in workflow or features.\n\nv1.1.0 | 2026-05-01T07:56:54.883Z | auto\n\n- Added support and documentation for `dlazy login` as the recommended authentication method for easier API key setup.\n- Updated authentication section to emphasize device code login workflow and simplified key management.\n- Clarified manual API key setup as an alternative, including updated instructions.\n- No changes to core workflow or platform support; all design and execution instructions remain the same.\n\nv1.0.9 | 2026-04-30T10:22:51.218Z | user\n\nSync to CLI 1.0.9 with cliWarning support\n\nv1.0.5 | 2026-04-29T02:35:55.872Z | user\n\nbump @dlazy/cli to 1.0.8\n\nv1.0.4 | 2026-04-27T08:03:57.654Z | user\n\nReduce false-positive scanner alerts: drop 'plaintext' wording from API key storage docs; remove persistsApiKey/network metadata flags in favour of neutral configLocation/apiEndpoints; rewrite Data & Privacy section as factual How-It-Works description without alarming warnings; emphasise that keys can be rotated/revoked at any time from the dLazy dashboard.\n\nv1.0.3 | 2026-04-27T07:42:41.036Z | user\n\nAdd provenance metadata (homepage/source/author/npm), document API key storage location (~/.dlazy/config.json) and DLAZY_API_KEY env var alternative, add Data & Privacy section, recommend 'npx @dlazy/cli@1.0.6' install alternative, normalise Chinese auth-error instruction wording.\n\nv1.0.2 | 2026-04-24T04:09:45.392Z | auto\n\n- Updated documentation files (SKILL.md, SKILL-en.md, SKILL-cn.md) with minor editorial adjustments.\n- No functional or logic changes; version bump for doc clarity and alignment.\n- All usage instructions and workflow remain unchanged.\n\nv1.0.1 | 2026-04-24T03:36:09.402Z | auto\n\n- 新增了身份验证说明，要求所有请求配置 dLazy API key，并详细介绍了获取和设置方法\n- dLazy CLI 依赖从 1.0.5 升级为 1.0.6\n- 文档结构优化，增加“身份验证 (Authentication)”独立章节，便于用户操作\n\nv1.0.0 | 2026-04-23T08:01:31.951Z | auto\n\nimage-social-media v1.0.0\n\n- Initial release of a structured skill for cross-platform social media content creation (Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, etc.).\n- Guides users through platform adaptation, visual strategy, content structuring, and layered copywriting.\n- Implements a strict step-by-step workflow requiring explicit user confirmation at every stage.\n- Enforces platform-specific technical/visual constraints and ensures safe-zone compliance.\n- Supports iterative design, multi-platform adaptation, and variant generation based on user feedback.\n- Provides critical execution rules for CLI use, enforcing single-command, synchronous executions without UI overlays or batch processing.\n\nArchive index:\n\nArchive v1.3.27: 4 files, 12016 bytes\n\nFiles: skill-card.md (1969b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nFile v1.3.27:SKILL.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nFile v1.3.27:_meta.json\n\n{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.27\",\n  \"publishedAt\": 1791422115591\n}\n\nFile v1.3.27:skill-card.md\n\n## Description:\n\nHelps create platform-tailored social-media visuals and captions, with plans for layouts, safe areas, and image generation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nCreators and marketing teams use this skill to plan and generate social-media images, thumbnails, and carousels adapted to each platform, with separate on-image copy and captions.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Prompts, generation settings, and media supplied for image generation may be sent to dLazy's hosted service.\n\nMitigation: Review material before submission and avoid sending sensitive or confidential content without authorization.\n\nRisk: Using the service requires an API key, and a global CLI installation persists locally.\n\nMitigation: Protect the key, rotate or revoke it in the dLazy dashboard when needed, and use npx instead of a global install if persistence is unwanted.\n\n## Reference(s):\n\n- [ClawHub skill release](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n- [dLazy CLI npm package](https://www.npmjs.com/package/@dlazy/cli)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, Guidance, Shell commands, Image URLs]\n\n**Output Format:** [Markdown with layout plans, copy, status checks, and optional generation commands or image URLs]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Includes platform specifications, safe-area checks, and suggested revisions; image generation uses the dLazy hosted service.]\n\n## Skill Version(s):\n\n1.3.27 (source: server-resolved release metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.3.27:SKILL-cn.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nArchive v1.3.26: 4 files, 12041 bytes\n\nFiles: skill-card.md (2023b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nFile v1.3.26:SKILL.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nFile v1.3.26:_meta.json\n\n{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.26\",\n  \"publishedAt\": 1791076444804\n}\n\nFile v1.3.26:skill-card.md\n\n## Description:\n\nGuides platform-specific social media image design, copywriting, and image generation through the dLazy CLI.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nCreators and marketing teams use this skill to plan and generate social media visuals, image text, and captions tailored to Instagram, TikTok, YouTube, LinkedIn, and Xiaohongshu.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Prompts, generation parameters, and supplied media may be sent to dLazy cloud services.\n\nMitigation: Avoid supplying sensitive content unless authorized to share it with dLazy.\n\nRisk: The dLazy CLI may save an API key locally for reuse.\n\nMitigation: Protect the local configuration and rotate or revoke the key if exposed.\n\nRisk: A global CLI installation persists on the local system.\n\nMitigation: Use the documented npx option or review the package before a global install.\n\n## Reference(s):\n\n- [ClawHub skill release](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n- [dLazy CLI package](https://www.npmjs.com/package/@dlazy/cli)\n- [dLazy CLI repository (referenced by skill metadata)](https://github.com/dlazy-ai/cli)\n- [dLazy service](https://dlazy.com)\n\n## Skill Output:\n\n**Output Type(s):** [Markdown, Guidance, Shell commands]\n\n**Output Format:** [Markdown text with design plans, captions, optional CLI commands, and generated image links]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Platform specifications, safe-area checks, and revision suggestions]\n\n## Skill Version(s):\n\n1.3.26 (source: server-resolved ClawHub release; skill frontmatter lists 1.3.7)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.3.26:SKILL-cn.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nArchive v1.3.25: 4 files, 12029 bytes\n\nFiles: skill-card.md (1978b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nFile v1.3.25:SKILL.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nFile v1.3.25:_meta.json\n\n{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.25\",\n  \"publishedAt\": 1790916994823\n}\n\nFile v1.3.25:skill-card.md\n\n## Description:\n\nGuides agents through planning and creating platform-tailored social-media visuals, captions, and image-generation requests.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nCreators and marketing teams use this skill to plan social-media images, thumbnails, and carousels for specific platforms, then generate visuals and separate caption copy.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Prompts and media supplied for generation may be sent to dLazy's cloud service.\n\nMitigation: Only provide content and media you are comfortable sending to dLazy.\n\nRisk: Signing in stores a dLazy API key locally.\n\nMitigation: Protect the local key and rotate or revoke it in your dLazy account if needed.\n\nRisk: A global CLI installation remains on the user's system.\n\nMitigation: Use the documented npx invocation for temporary use if a global install is not desired.\n\n## Reference(s):\n\n- [ClawHub skill listing](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n- [dLazy CLI package](https://www.npmjs.com/package/@dlazy/cli)\n- [dLazy CLI source repository](https://github.com/dlazy-ai/cli)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Text, Shell commands, Image URLs]\n\n**Output Format:** [Markdown with design plans, caption copy, CLI commands, and generated image links]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Keeps in-image text separate from captions and checks platform dimensions and safe areas.]\n\n## Skill Version(s):\n\n1.3.25 (source: ClawHub release evidence; artifact frontmatter says 1.3.7)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.3.25:SKILL-cn.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nArchive v1.3.24: 4 files, 11998 bytes\n\nFiles: skill-card.md (1917b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nFile v1.3.24:SKILL.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nFile v1.3.24:_meta.json\n\n{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.24\",\n  \"publishedAt\": 1790730777054\n}\n\nFile v1.3.24:skill-card.md\n\n## Description:\n\nPlans and generates platform-specific social media visuals and captions for Instagram, TikTok, YouTube, LinkedIn, and Xiaohongshu.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nCreators and marketing teams use this skill to plan, generate, and revise social media graphics, carousel content, thumbnails, and accompanying captions for specific platforms.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Prompts and supplied media may be sent to dLazy's cloud service, and generated images are hosted there.\n\nMitigation: Avoid submitting sensitive prompts or local files unless cloud processing and hosting are intended.\n\nRisk: The third-party CLI may store an API key locally and introduces package supply-chain exposure.\n\nMitigation: Review the CLI package before installation and protect or rotate locally stored API keys.\n\n## Reference(s):\n\n- [ClawHub skill listing](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n- [dLazy CLI package](https://www.npmjs.com/package/@dlazy/cli)\n- [dLazy service](https://dlazy.com)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, Shell commands, Image URLs]\n\n**Output Format:** [Markdown with design plans, image links, and separate caption copy]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Includes platform specifications, safe-area checks, generation status, and iteration suggestions; generated image URLs are hosted by dLazy.]\n\n## Skill Version(s):\n\n1.3.24 (source: ClawHub release metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.3.24:SKILL-cn.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nArchive v1.3.23: 4 files, 11879 bytes\n\nFiles: skill-card.md (1655b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nFile v1.3.23:SKILL.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nFile v1.3.23:_meta.json\n\n{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.23\",\n  \"publishedAt\": 1790561516162\n}\n\nFile v1.3.23:skill-card.md\n\n## Description:\n\nPlans platform-specific social media visuals and captions and supports image generation through the dLazy CLI.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nCreators and marketing teams use this skill to plan and generate social media graphics, platform-specific layouts, in-image headlines, and accompanying captions.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The third-party CLI uses a dLazy API key.\n\nMitigation: Review the CLI before use, protect the key, and rotate or revoke it if needed.\n\nRisk: Prompts and explicitly supplied media files are sent to dLazy's hosted service.\n\nMitigation: Only supply prompts and media approved for upload to that service.\n\n## Reference(s):\n\n- [ClawHub skill listing](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n- [dLazy CLI package](https://www.npmjs.com/package/@dlazy/cli)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Shell commands, Image URLs]\n\n**Output Format:** [Markdown with design plans, captions, and generated image links]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Includes platform specifications, safe-area checks, and suggested revisions.]\n\n## Skill Version(s):\n\n1.3.23 (source: server-resolved ClawHub release)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.3.23:SKILL-cn.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nArchive v1.3.22: 4 files, 12109 bytes\n\nFiles: skill-card.md (2144b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nFile v1.3.22:SKILL.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nFile v1.3.22:_meta.json\n\n{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.22\",\n  \"publishedAt\": 1790558035461\n}\n\nFile v1.3.22:skill-card.md\n\n## Description:\n\nPlans and creates platform-adapted social-media visuals, in-image text, and captions for Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and other channels.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nCreators and marketing teams use this skill to plan and generate social-media imagery and captions tailored to each platform's format, visual style, and engagement goal.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Prompts and any media supplied for generation are sent to dLazy's cloud service.\n\nMitigation: Share only material you are comfortable sending to dLazy and review media before submission.\n\nRisk: Authentication can store an API key locally.\n\nMitigation: Use an environment variable if you prefer not to save a key; rotate or revoke credentials when needed.\n\nRisk: The skill relies on a third-party CLI installation.\n\nMitigation: Confirm the package and version before use, and consider the on-demand npx option instead of a global install.\n\n## Reference(s):\n\n- [ClawHub skill release](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n- [dLazy CLI package](https://www.npmjs.com/package/@dlazy/cli)\n- [dLazy CLI documentation and code](https://github.com/dlazy-ai/cli)\n- [dLazy website](https://dlazy.com)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, Image URLs, Design guidance]\n\n**Output Format:** [Markdown with platform-specific layout plans, in-image copy, captions, and links to generated images]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Includes aspect-ratio and safe-area checks; keeps captions separate from text inside images.]\n\n## Skill Version(s):\n\n1.3.22 (source: ClawHub release; supplied skill frontmatter lists 1.3.7)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.3.22:SKILL-cn.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nArchive v1.3.21: 4 files, 12183 bytes\n\nFiles: skill-card.md (2338b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nFile v1.3.21:SKILL.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nFile v1.3.21:_meta.json\n\n{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.21\",\n  \"publishedAt\": 1790232589982\n}\n\nFile v1.3.21:skill-card.md\n\n## Description:\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and other formats with platform-specific layout, copy, safe-area, and generation guidance.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal creators, marketers, and agents use this skill to plan and generate social-media visuals and accompanying captions for platform-specific posts, carousels, stories, shorts, thumbnails, and professional content.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Prompts and media file paths supplied for generation may be sent to dLazy-hosted endpoints.\n\nMitigation: Avoid submitting confidential, sensitive, or unapproved media, and review dLazy's service expectations before using hosted generation.\n\nRisk: Authentication stores a dLazy API key in the user's local CLI configuration.\n\nMitigation: Use an organization-scoped key, rotate or revoke it when needed, and prefer per-command environment variables when persistent local credentials are not desired.\n\nRisk: Global CLI installation creates a long-lived dependency on the local machine.\n\nMitigation: Use the documented npx command for ad hoc execution when a persistent global install is not needed.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n- [dLazy CLI source](https://github.com/dlazy-ai/cli)\n- [dLazy CLI npm package](https://www.npmjs.com/package/@dlazy/cli)\n- [dLazy homepage](https://dlazy.com)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown with structured planning, copy, checks, and inline shell commands]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May return hosted image result URLs from the dLazy service after CLI execution.]\n\n## Skill Version(s):\n\n1.3.21 (source: server release metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.3.21:SKILL-cn.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nArchive v1.3.20: 4 files, 12213 bytes\n\nFiles: skill-card.md (2463b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nFile v1.3.20:SKILL.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-second rule:\n\n- Bold claim: e.g., \"Stop doing X\"\n- Curiosity gap: e.g., \"The secret to ...\"\n- Visual impact: unconventional palette or composition\n\nCarousel structure must follow:\n\n- Page 1: hook (why keep swiping)\n- Page 2: value reinforcement\n- Pages 3–N: core content blocks\n- Last page: explicit CTA\n\n### Phase 3: Generation Constraints\n\nValidate these hard constraints before generating:\n\n- Aspect ratio is correct, and all carousel pages share the same ratio\n- Key elements sit inside the central safe area\n- Text has high readability (shadow, gradient, contrast)\n- The user-specified color leads the palette\n- Do not render like buttons, duration markers, resolution badges, or other platform UI\n- Avoid a flat-sticker feel; preserve hierarchy, volume, or lighting depth\n\n### Phase 4: Layered Copy Output\n\nText must be layered:\n\n- In-image text: short headline, kept under 10 words\n- Caption: long copy, returned separately, never rendered into the image\n\nWhen the user does not provide caption copy, auto-generate per platform:\n\n- Instagram: short sentences + line breaks + 3–5 hashtags\n- TikTok: 1 line of high-impact copy\n- YouTube: SEO structure + timestamps + subscribe CTA\n- LinkedIn: value-first + bullet points + professional CTA\n- Xiaohongshu: dense emoji + colloquial social tone\n\n### Phase 5: Iteration and Expansion\n\nOnce the user is satisfied, proactively offer next moves:\n\n- Style variants on the same theme\n- Multi-platform adaptation of the same content\n- Expand a single image into a carousel\n- Add or improve caption copy\n\nWhen the user is unsatisfied, prioritize:\n\n- Color adjustments\n- Composition and whitespace re-balance\n- Style switch\n- Platform switch with new aspect ratio and safe area\n- Hook reinforcement\n\n## Platform Specs and Visual Style\n\n| Platform | Format | Visual Style |\n| --- | --- | --- |\n| Instagram | Feed 1:1 or 4:5 (recommended) / Story, Reels 9:16 | Polished, poster-like, high resolution |\n| TikTok / YouTube Shorts | 9:16, subject centered or left | Authentic, high-energy, native feel, text-dense |\n| LinkedIn | 4:5 or PDF carousel | Clean, corporate, infographic, blue-gray palette |\n| YouTube thumbnail | 16:9 | High contrast, exaggerated expressions, big-title (≤5 words) |\n| Xiaohongshu | 3:4 | Collage style, heavy emoji, headline overlay |\n| Carousel general | Up to 20 pages, all same ratio | Series-level consistency first |\n\n## Safe Area and Whitespace Rules\n\nKey elements (text, faces, products) must stay inside the central safe area and avoid platform overlays:\n\n- Instagram Story / Reels: top 15%, bottom 20%\n- TikTok / Shorts: right-side interaction strip and bottom caption strip\n- YouTube thumbnail: bottom-right duration zone\n\nWhitespace and subject share rules:\n\n- Standard whitespace: at least 15% from the edges\n- Premium feel: up to 40% negative space\n- Subject zone: core elements occupy the center 60% of the visual area\n\n## Algorithm Goal Mapping\n\n- Goal = share: high contrast, meme grammar, strong-resonance phrasing\n- Goal = save: infographic structure, step breakdowns, listicles\n- Goal = comment: A/B comparisons, provocative questions, open-ended conclusions\n\n## Output Format\n\nEach output includes:\n\n- Current phase and task status\n- Platform spec and safe-area check results\n- Layout and copy plan\n- Delivered content (in-image text + caption)\n- Next-step suggestions (variant / adaptation / iteration)\n\n\n## 🛠️ 执行与生成指南 (CRITICAL EXECUTION INSTRUCTIONS)\n\n**你是可以执行终端命令的智能 Agent！**\n\n**【严格禁止行为】**\n- 严禁：将提示词保存到任何文件中（如 txt, md）。\n- 严禁：要求用户自己去第三方平台（如 Midjourney）生成图片。\n- 严禁：一次性批量生成所有图片，或一次性执行多个命令。\n\n**【必须遵循的交互与执行流程】**\n你必须**严格分步**执行，并在每一步停下来等待用户回复：\n\n1. **第一步：主动收集需求**。当用户提出需求时，不要做任何设计和生成，先向用户提问（如产品特点、目标人群、想要几张图等）。**必须等待用户回答。**\n2. **第二步：输出草案并请求确认**。根据用户的回答，制定套图计划，并输出**第一张图**的提示词草案。**询问用户：“是否确认这个提示词，可以开始生成第一张图了吗？” 必须等待用户回答“确认”。**\n3. **第三步：单次执行终端命令**。用户确认后，你**必须使用终端执行命令**（如 `dlazy seedream-4.5 --prompt \"...\"`），每次只能执行一个生成命令。**重要：必须使用同步命令，绝不要在命令末尾加 `&`，绝不要使用 `&&`，这是在 Windows PowerShell 下运行！**\n4. **第四步：交付与循环**。命令返回结果后，把图片 URL 发给用户，并询问“对这张满意吗？我们可以继续生成下一张了吗？”。收到确认后再继续下一步。\n\nFile v1.3.20:_meta.json\n\n{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.20\",\n  \"publishedAt\": 1790039497935\n}\n\nFile v1.3.20:skill-card.md\n\n## Description:\n\nA structured skill for creating multi-platform social media visuals and captions for Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and related formats.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal creators, marketers, and agent users use this skill to plan, generate, and iterate social media images, thumbnails, carousel structures, in-image text, and platform-specific captions.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Prompts, parameters, and selected media files may be sent to dLazy-hosted services during generation.\n\nMitigation: Avoid providing sensitive local media or confidential prompts unless upload to dLazy services is intended.\n\nRisk: Using a persistent global CLI install or saved API key can leave local credentials and tooling in place after use.\n\nMitigation: Prefer npx or the DLAZY_API_KEY environment variable for temporary use, and rotate or revoke API keys from the dLazy dashboard when needed.\n\nRisk: Generated social media assets can be off-brand, platform-inaccurate, or include unsuitable caption or in-image text.\n\nMitigation: Review generated layouts, safe-area checks, image text, and captions before publication.\n\n## Reference(s):\n\n- [dLazy CLI source repository](https://github.com/dlazy-ai/cli)\n- [dLazy CLI npm package](https://www.npmjs.com/package/@dlazy/cli)\n- [dLazy website](https://dlazy.com)\n- [dLazy API key dashboard](https://dlazy.com/dashboard/organization/api-key)\n- [ClawHub skill page](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown with planning notes, safe-area checks, copy layers, captions, inline shell commands, and generated asset URLs.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May call the dLazy CLI to submit prompts, parameters, and selected media files to dLazy-hosted services when generation is confirmed.]\n\n## Skill Version(s):\n\n1.3.20 (source: server release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.3.20:SKILL-cn.md\n\n---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `write_todos` tool to set up a task plan that includes at least:\n\n- Goal and platform-spec confirmation\n- Hook and content-structure planning\n- Visual generation and quality check\n- Variant or multi-platform iteration\n\nExecution rules:\n\n- Keep only one task `in_progress`; the rest are `pending`.\n- Update `write_todos` status as soon as each phase finishes.\n- When the user asks for revisions or platform switches, add or re-order tasks and continue.\n\n### Phase 1: Goal and Platform Definition\n\nYou must clarify first:\n\n1. Publishing platform and format\n2. Engagement goal (share / save / comment)\n3. Content form (single image / carousel / thumbnail)\n4. Target audience and tone\n\nIf the user's input is incomplete, prioritize completing platform and goal before moving on.\n\n### Phase 2: Hook and Structure Planning\n\nDesign first-frame attention via the 3-se\n\nArchive v1.3.19: 4 files, 12263 bytes\n\nFiles: skill-card.md (2477b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)\n\nArchive v1.3.18: 4 files, 12207 bytes\n\nFiles: skill-card.md (2447b), SKILL-cn.md (9829b), SKILL.md (9829b), _meta.json (144b)","readmeExcerpt":"Skill: 社交媒体配图设计 Social Media Design Owner: dlazyai Summary: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs... Tags: latest:1.3.27 Version history: v1.3.27 | 2026-10-08T01:15:15.591Z | user 例行版本更新 2026-10-08 v1.3.26 | 2026-10-04T01:14:04.804Z | user 例行版本更新 2026-10-04 v1.3.25 | 2026-10-02T04:56:34.823Z | user 例","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"dlazy login"},{"language":"bash","snippet":"dlazy auth set YOUR_API_KEY"},{"language":"bash","snippet":"npx @dlazy/cli@1.2.3 <command>"},{"language":"bash","snippet":"dlazy login"},{"language":"bash","snippet":"dlazy auth set YOUR_API_KEY"},{"language":"bash","snippet":"npx @dlazy/cli@1.2.3 <command>"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `writ"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn7c5wgeajfcfvdfb5ceemvdb984cjpd\",\n  \"slug\": \"dlazy-image-social-media\",\n  \"version\": \"1.3.27\",\n  \"publishedAt\": 1791422115591\n}"},{"path":"skill-card.md","content":"## Description:\n\nHelps create platform-tailored social-media visuals and captions, with plans for layouts, safe areas, and image generation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[dlazyai](https://clawhub.ai/user/dlazyai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nCreators and marketing teams use this skill to plan and generate social-media images, thumbnails, and carousels adapted to each platform, with separate on-image copy and captions.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Prompts, generation settings, and media supplied for image generation may be sent to dLazy's hosted service.\n\nMitigation: Review material before submission and avoid sending sensitive or confidential content without authorization.\n\nRisk: Using the service requires an API key, and a global CLI installation persists locally.\n\nMitigation: Protect the key, rotate or revoke it in the dLazy dashboard when needed, and use npx instead of a global install if persistence is unwanted.\n\n## Reference(s):\n\n- [ClawHub skill release](https://clawhub.ai/dlazyai/skills/dlazy-image-social-media)\n- [dLazy CLI npm package](https://www.npmjs.com/package/@dlazy/cli)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, Guidance, Shell commands, Image URLs]\n\n**Output Format:** [Markdown with layout plans, copy, status checks, and optional generation commands or image URLs]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Includes platform specifications, safe-area checks, and suggested revisions; image generation uses the dLazy hosted service.]\n\n## Skill Version(s):\n\n1.3.27 (source: server-resolved release metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment."},{"path":"SKILL-cn.md","content":"---\nname: dlazy-image-social-media\nversion: 1.3.7\ndescription: A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati\ntriggers:\n  - Social Media Designer (Multi-Platform Optimization)\nmetadata: {\"clawdbot\":{\"emoji\":\"🤖\",\"requires\":{\"bins\":[\"npm\",\"npx\"]},\"install\":\"npm install -g @dlazy/cli@1.2.3\",\"installAlternative\":\"npx @dlazy/cli@1.2.3\",\"homepage\":\"https://github.com/dlazy-ai/cli\",\"source\":\"https://github.com/dlazy-ai/cli\",\"author\":\"dlazyai\",\"license\":\"see-repo\",\"npm\":\"https://www.npmjs.com/package/@dlazy/cli\",\"configLocation\":\"~/.dlazy/config.json\",\"apiEndpoints\":[\"api.dlazy.com\",\"files.dlazy.com\"]},\"openclaw\":{\"systemPrompt\":\"当你需要使用此技能时，请严格遵循此技能提供的指南进行规划和执行。你可以通过调用 dlazy CLI 的各类生成模型（如 dlazy seedream-4.5 等）来完成实际的图片渲染。注意：Windows PowerShell 中不允许使用 `&` 或 `&&` 进行命令串联或后台运行，请单独且同步地执行命令。\"}}\n---\n\n## 身份验证 (Authentication)\n\n所有请求都需要 dLazy API key。**推荐使用** `dlazy login` 完成登录：\n\n```bash\ndlazy login\n```\n\n该命令使用设备码流程（远程终端也可用），登录成功后 **自动把 API key 写入本地 CLI 配置**，无需手动复制粘贴。\n\n### 备选：手动设置 API Key\n\n如果你已有 API key，也可以直接保存：\n\n```bash\ndlazy auth set YOUR_API_KEY\n```\n\nCLI 会把 key 保存在你的用户配置目录（macOS/Linux 上为 `~/.dlazy/config.json`，Windows 上为 `%USERPROFILE%\\.dlazy\\config.json`），文件权限仅限当前操作系统用户访问。你也可以用 `DLAZY_API_KEY` 环境变量按次传入。\n\n### 手动获取 API Key\n\n1. 登录或在 [dlazy.com](https://dlazy.com) 创建账号\n2. 访问 [dlazy.com/dashboard/organization/api-key](https://dlazy.com/dashboard/organization/api-key)\n3. 复制 API Key 区域显示的密钥\n\n每个 key 都属于你自己的 dLazy 组织，可在同一控制面板**随时轮换或吊销**。\n\n## 关于与来源 (Provenance)\n\n- **CLI 源代码**: [github.com/dlazy-ai/cli](https://github.com/dlazy-ai/cli)\n- **维护者**: dlazyai\n- **npm 包名**: `@dlazy/cli`（本技能 install 字段固定到 `1.0.9` 版本）\n- **官网**: [dlazy.com](https://dlazy.com)\n\n如果你不希望在系统上长期保留一个全局 CLI，可以按需运行：\n\n```bash\nnpx @dlazy/cli@1.2.3 <command>\n```\n\n如选择全局安装，技能的 `metadata.clawdbot.install` 字段已固定到 `npm install -g @dlazy/cli@1.2.3`。安装前建议先到 GitHub 仓库审阅源码。\n\n## 工作原理 (How It Works)\n\n此技能是 dLazy 托管 API 的轻量封装。调用时：\n\n- 你提供的提示词与参数会发送到 dLazy API（`api.dlazy.com`）进行推理。\n- 传入图像 / 视频 / 音频字段的本地文件路径会被 CLI 上传到 dLazy 媒体存储（`files.dlazy.com`），以便模型读取 —— 与任何云端生成 API 的流程一致。\n- API 返回的生成结果 URL 由 `files.dlazy.com` 托管。\n\n这是标准的 SaaS 调用模式；技能本身不会越权访问网络或文件系统，所有动作都由 dLazy CLI 完成。\n\n# 社交媒体配图设计 Social Media Design\n\n[English](./SKILL.md) · [中文](./SKILL-cn.md)\n\nA structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's native expectations across technical specs, visual language, and engagement strategy.\n\n## Core Positioning\n\nYour responsibilities:\n\n- ✅ Platform adaptation strategy and visual decisions\n- ✅ Executable layout and content-structure planning\n- ✅ Layered output of in-image text and caption copy\n- ❌ Rendering platform UI elements or unnecessary technical noise\n\n## Execution Framework\n\n### Step 0: Task Planning (Mandatory)\n\nBefore any design output, call the `writ"}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":null,"editorialQuality":{"score":100,"threshold":65,"status":"thin","wordCount":1244,"uniquenessScore":41,"reasons":["uniqueness-below-45"]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-09T04:52:37.536Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-09T04:52:37.536Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-10T10:46:10.192Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}