{"id":"0505da5a-ff10-4280-ad48-a48ea66e16a3","entityType":"agent","slug":"clawhub-cool131219-reply-wechat-message","name":"Reply Wechat Message","canonicalUrl":"https://www.xpersona.co/agent/clawhub-cool131219-reply-wechat-message","canonicalPath":"/agent/clawhub-cool131219-reply-wechat-message","generatedAt":"2026-10-10T11:50:41.018Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-10T09:40:50.346Z","emptyReason":null},"description":"微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply Skill: Reply Wechat Message Owner: cool131219 Summary: 微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply Tags: latest:1.0.4 Version history: v1.0.4 | 2026-07-09T08:30:40.963Z | auto - No code or documentation changes detected in this release. - Version increment only; functionality remains unchanged. v1.0.3 | 2026-07-09T08:27:48.728Z | auto - No file changes detected in this release. - Version","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.5K downloads reported by the source. Last updated 10/10/2026.","installCommand":"clawhub skill install s176tzyws7adkp21qbg139qm2x87n4vv:reply-wechat-message","sourceUrl":"https://clawhub.ai/cool131219/reply-wechat-message","homepage":"https://clawhub.ai/cool131219/skills/reply-wechat-message","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/cool131219/reply-wechat-message","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/cool131219/skills/reply-wechat-message","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":64,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply Skill: Reply Wechat Message Owner: cool131219 Summary: 微信AI助手 — 主动发消息 + 自动回复 / WeChat A"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-10T09:40:50.346Z","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":"2026-10-10T09:40:50.346Z","emptyReason":null},"stars":null,"forks":null,"downloads":1519,"packageName":null,"latestVersion":"1.0.4","tractionLabel":"1.5K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-10T09:40:50.346Z","emptyReason":null},"lastUpdatedAt":"2026-10-10T09:40:50.346Z","lastCrawledAt":"2026-10-10T09:40:50.346Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-11T09:40:50.346Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.4","createdAt":"2026-07-09T08:30:40.963Z","changelog":"- No code or documentation changes detected in this release. - Version increment only; functionality remains unchanged.","fileCount":3,"zipByteSize":4262},{"version":"1.0.3","createdAt":"2026-07-09T08:27:48.728Z","changelog":"- No file changes detected in this release. - Version metadata updated; functionality and documentation remain unchanged.","fileCount":3,"zipByteSize":4253},{"version":"1.0.2","createdAt":"2026-07-09T08:26:58.673Z","changelog":"- Updated metadata fields in SKILL.md, including new fields: slug, displayName, and agent_created. - Removed the skill-card.md file. - Simplified SKILL.md by removing detailed metadata and dependencies sections. - No changes to skill functionality or usage.","fileCount":3,"zipByteSize":4243},{"version":"1.0.1","createdAt":"2026-07-09T04:58:34.503Z","changelog":"- Added proactive message sending capability; now supports sending messages directly to contacts. - All related documentation updated and consolidated into SKILL.md; installation and individual script docs removed. - Triggers, commands, and usage instructions are now fully bilingual (Chinese & English). - Reduced skill footprint by removing redundant files. - Clarified and streamlined messaging and auto-reply workflow.","fileCount":3,"zipByteSize":4480},{"version":"1.0.0","createdAt":"2026-07-09T00:19:33.525Z","changelog":"- Initial release: Automatically replies to WeChat contacts using AI. - Reads the entire chat context (including sender distinctions), analyzes conversation flow, and generates smart replies. - Fully self-contained: includes scripts for launching WeChat and sending messages (no extra skills required). - Requires only Python (plus common pip packages); no separate OCR registration necessary. - Multiple trigger formats supported, including both concise and natural language commands.","fileCount":8,"zipByteSize":12007}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s176tzyws7adkp21qbg139qm2x87n4vv:reply-wechat-message","setupComplexity":"low","setupSteps":["Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"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-cool131219-reply-wechat-message/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T11:50:41.017Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cool131219-reply-wechat-message/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":"high","updatedAt":"2026-10-10T09:40:50.346Z","emptyReason":null},"readme":"Skill: Reply Wechat Message\n\nOwner: cool131219\n\nSummary: 微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply\n\nTags: latest:1.0.4\n\nVersion history:\n\nv1.0.4 | 2026-07-09T08:30:40.963Z | auto\n\n- No code or documentation changes detected in this release.\n- Version increment only; functionality remains unchanged.\n\nv1.0.3 | 2026-07-09T08:27:48.728Z | auto\n\n- No file changes detected in this release.\n- Version metadata updated; functionality and documentation remain unchanged.\n\nv1.0.2 | 2026-07-09T08:26:58.673Z | auto\n\n- Updated metadata fields in SKILL.md, including new fields: slug, displayName, and agent_created.\n- Removed the skill-card.md file.\n- Simplified SKILL.md by removing detailed metadata and dependencies sections.\n- No changes to skill functionality or usage.\n\nv1.0.1 | 2026-07-09T04:58:34.503Z | auto\n\n- Added proactive message sending capability; now supports sending messages directly to contacts.\n- All related documentation updated and consolidated into SKILL.md; installation and individual script docs removed.\n- Triggers, commands, and usage instructions are now fully bilingual (Chinese & English).\n- Reduced skill footprint by removing redundant files.\n- Clarified and streamlined messaging and auto-reply workflow.\n\nv1.0.0 | 2026-07-09T00:19:33.525Z | auto\n\n- Initial release: Automatically replies to WeChat contacts using AI.\n- Reads the entire chat context (including sender distinctions), analyzes conversation flow, and generates smart replies.\n- Fully self-contained: includes scripts for launching WeChat and sending messages (no extra skills required).\n- Requires only Python (plus common pip packages); no separate OCR registration necessary.\n- Multiple trigger formats supported, including both concise and natural language commands.\n\nArchive index:\n\nArchive v1.0.4: 3 files, 4262 bytes\n\nFiles: skill-card.md (2040b), SKILL.md (6185b), _meta.json (139b)\n\nFile v1.0.4:SKILL.md\n\n---\nname: reply-wechat-message\nslug: wechat-butler\ndisplayName: 微信管家 / WeChat Butler\ndescription: \"微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply\"\nagent_created: true\n---\n\n# 微信管家 / WeChat Butler\n\n微信AI助手 — 主动发消息给联系人 + 收到消息后AI自动回复。\nWeChat AI assistant — proactively send messages, and auto-reply when messages come in.\n\n**自包含技能包** — 所有依赖脚本（启动微信、发送消息）已打包在内，无需额外安装。\n**Self-contained skill** — all dependency scripts (launch WeChat, send messages) are bundled. No extra installation needed.\n\n---\n\n## 功能一：AI 主动发消息 / Feature 1: Send Message\n\n直接发送一条消息给指定联系人（不读取聊天上下文，即时发送）。\nSend a message directly to a contact (no context reading, instant send).\n\n### 触发格式 / Trigger Format\n\n**中文 / Chinese:**\n```\n给 [联系人] 发消息：[内容]\n发消息给 [联系人]：[内容]\n帮 [联系人] 发消息：[内容]\n```\n\n**English:**\n```\nsend [contact] [message]\nmsg [contact] [message]\nmessage [contact] [message]\n```\n\n### 示例 / Examples\n\n**中文 / Chinese:**\n- `给 小明 发消息：中午一起去吃饭吗？`\n- `发消息给 小红：记得带文件`\n- `帮 小张 发消息：生日快乐！`\n\n**English:**\n- `send Kitty: Want to grab lunch?`\n- `msg Peter: Don't forget the documents`\n- `message Tom: Happy birthday!`\n\n### AI 执行步骤 / Execution Steps\n\n```\n第1步 / Step 1:\n AI 提取联系人和消息内容\n AI extracts contact name and message content\n\n第2步 / Step 2:\n python scripts/send_wechat.py <联系人/contact> <内容/message>\n → 打开微信 → 搜索联系人 → 打开聊天 → 发送消息\n → Open WeChat → search contact → open chat → send message\n\n第3步 / Step 3:\n 返回发送结果给用户\n Report send result to user\n```\n\n---\n\n## 功能二：AI 自动回复 / Feature 2: AI Auto-Reply\n\n读取聊天上下文，AI 分析后自动回复。\nReads the chat context, AI analyzes it, then auto-replies.\n\n### 触发格式 / Trigger Format\n\n```\n使用技能:reply-wechat-message 给：[联系人] 回复\n```\n\n### 示例 / Examples\n- `使用技能:reply-wechat-message 给：小明 回复` (Xiao Ming / Kitty)\n- `使用技能:reply-wechat-message 给：小红 回复` (Xiao Hong / Peter)\n\n### 简化触发 / Short Forms\n- 回复 小明 / reply Kitty\n- 回 小红 / reply Peter\n\n---\n\n## AI 工作流程（自动回复）/ AI Workflow (Auto-Reply)\n\n```\n第1步 / Step 1:\n python scripts/reply_wechat.py 小明 / python scripts/reply_wechat.py Kitty\n → 读取整个聊天区域，区分左右两侧\n → Read entire chat area, split left/right\n → 输出结构化对话 / Output structured conversation:\n ★上下文开始★ / ★Context Start★\n [对方/Them] xxx\n [我/Me] xxx\n [对方/Them] xxx\n ★上下文结束★ / ★Context End★\n\n第2步 / Step 2:\n AI分析全部上下文，理解对话脉络，生成合适的回复\n AI analyzes full context, understands the conversation flow, generates a suitable reply\n\n ⚠️ 重要规则 / Important Rule:\n OCR识别出的聊天内容即为真实上下文，模型不得质疑、猜测或怀疑识别结果的准确性。\n The OCR-recognized text IS the real context. The model must NOT question, guess,\n or doubt the accuracy of the recognition. Reply based on the recognized content directly.\n\n第3步 / Step 3:\n echo \"AI生成的回复 / AI-generated reply\" | python scripts/reply_wechat.py 小明\n → 或 / Or: echo \"AI-generated reply\" | python scripts/reply_wechat.py Kitty\n → 读取上下文 + 自动发送回复（一步完成，stdin管道无引号问题）\n → Read context + send reply in one step (stdin piping avoids quote issues)\n```\n\n---\n\n## 脚本 / Scripts\n\n`scripts\\send_wechat.py` — 发送脚本：搜索联系人 + 发送消息 / Send script: search contact + send message\n`scripts\\reply_wechat.py` — 主脚本：读取上下文 + 发送回复 / Main script: read context + send reply\n`scripts\\open_wechat.py` — 启动脚本：唤醒微信窗口 / Launch script: bring WeChat to foreground\n\n### send_wechat.py 用法 / Usage\n```bash\n# 发送消息 / Send message (command line arg)\npython scripts/send_wechat.py <小明/Kitty> <消息/message>\n\n# 发送消息（stdin管道，无引号问题）/ Send message (stdin pipe, no quote issues)\necho \"消息/message\" | python scripts/send_wechat.py <小明/Kitty>\n```\n\n### reply_wechat.py 用法 / Usage\n```bash\n# 只读取上下文（AI分析用）/ Read context only (for AI analysis)\npython scripts/reply_wechat.py <小明/Kitty>\n\n# 读取上下文 + 发送回复（一步到位）/ Read context + send reply (one step)\necho \"回复内容/reply text\" | python scripts/reply_wechat.py <小明/Kitty>\n```\n\n### 输出格式 / Output Format\n```\n★上下文开始★ / ★Context Start★\n[对方/Them] 你吃饭了吗 / Have you eaten?\n[我/Me] 吃过了，你呢 / Yes, and you?\n[对方/Them] 我也吃了 / Me too\n★上下文结束★ / ★Context End★\n```\n\n---\n\n## 消息区分逻辑 / Message Detection Logic\n\n- **左侧（白底黑字）/ Left side (white bg, black text)** = 对方发的消息 / Messages from the other party → 标注 `[对方/Them]`\n- **右侧（绿底黑字）/ Right side (green bg, black text)** = 自己发的消息 / Messages from yourself → 标注 `[我/Me]`\n- 截图聊天区从中线切开，左右分别OCR，避免颜色误判\n Screenshot is split at the center line; left and right are OCR'd separately to avoid color misidentification\n\n---\n\n## 依赖 / Dependencies\n\n- Python 3.10+\n- OCR.space API（免费版，无需注册 / Free tier, no registration required）\n- Python packages: pyautogui, pygetwindow, pyperclip, requests, Pillow, numpy\n\n---\n\n## 脚本清单 / Script Inventory\n\n| 文件 / File | 功能 / Function |\n|---|---|\n| `send_wechat.py` | 搜索联系人 + 发送消息 / Search contact & send message |\n| `reply_wechat.py` | 读取上下文 + 发送回复 / Read context & send reply |\n| `open_wechat.py` | 启动微信 / Launch WeChat |\n\nFile v1.0.4:_meta.json\n\n{\n \"ownerId\": \"kn7022zkdvgpkbwdd7wc1vppex87m15m\",\n \"slug\": \"reply-wechat-message\",\n \"version\": \"1.0.4\",\n \"publishedAt\": 1783585840963\n}\n\nFile v1.0.4:skill-card.md\n\n## Description:\n\nWeChat AI assistant that sends messages to contacts and auto-replies based on visible chat context.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[cool131219](https://clawhub.ai/user/cool131219)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users and developers use this skill to ask an agent to send WeChat messages or read visible WeChat chat context and prepare auto-replies for a selected contact.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill can read visible WeChat conversations and may expose chat content to an external OCR service.\n\nMitigation: Use it only for non-sensitive chats unless explicit OCR consent, local OCR, or clear external OCR disclosure is added.\n\nRisk: The skill can send WeChat messages automatically as the user.\n\nMitigation: Require a human preview and confirmation step before any outbound message is sent.\n\nRisk: Security evidence reports unsafe shell examples and missing bundled scripts.\n\nMitigation: Review the supplied scripts before installation and replace unsafe shell usage with safer subprocess or input handling.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/cool131219/skills/reply-wechat-message)\n- [Server-resolved GitHub source](https://github.com/cool131219/WeChat-Butler/tree/main/reply-wechat-message)\n\n## Skill Output:\n\n**Output Type(s):** [text, shell commands, guidance]\n\n**Output Format:** [Markdown instructions with inline shell commands and structured conversation text]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May read visible WeChat chat context and send messages through bundled scripts when present.]\n\n## Skill Version(s):\n\n1.0.4 (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\nArchive v1.0.3: 3 files, 4253 bytes\n\nFiles: skill-card.md (2071b), SKILL.md (6185b), _meta.json (139b)\n\nFile v1.0.3:SKILL.md\n\n---\nname: reply-wechat-message\nslug: wechat-butler\ndisplayName: 微信管家 / WeChat Butler\ndescription: \"微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply\"\nagent_created: true\n---\n\n# 微信管家 / WeChat Butler\n\n微信AI助手 — 主动发消息给联系人 + 收到消息后AI自动回复。\nWeChat AI assistant — proactively send messages, and auto-reply when messages come in.\n\n**自包含技能包** — 所有依赖脚本（启动微信、发送消息）已打包在内，无需额外安装。\n**Self-contained skill** — all dependency scripts (launch WeChat, send messages) are bundled. No extra installation needed.\n\n---\n\n## 功能一：AI 主动发消息 / Feature 1: Send Message\n\n直接发送一条消息给指定联系人（不读取聊天上下文，即时发送）。\nSend a message directly to a contact (no context reading, instant send).\n\n### 触发格式 / Trigger Format\n\n**中文 / Chinese:**\n```\n给 [联系人] 发消息：[内容]\n发消息给 [联系人]：[内容]\n帮 [联系人] 发消息：[内容]\n```\n\n**English:**\n```\nsend [contact] [message]\nmsg [contact] [message]\nmessage [contact] [message]\n```\n\n### 示例 / Examples\n\n**中文 / Chinese:**\n- `给 小明 发消息：中午一起去吃饭吗？`\n- `发消息给 小红：记得带文件`\n- `帮 小张 发消息：生日快乐！`\n\n**English:**\n- `send Kitty: Want to grab lunch?`\n- `msg Peter: Don't forget the documents`\n- `message Tom: Happy birthday!`\n\n### AI 执行步骤 / Execution Steps\n\n```\n第1步 / Step 1:\n AI 提取联系人和消息内容\n AI extracts contact name and message content\n\n第2步 / Step 2:\n python scripts/send_wechat.py <联系人/contact> <内容/message>\n → 打开微信 → 搜索联系人 → 打开聊天 → 发送消息\n → Open WeChat → search contact → open chat → send message\n\n第3步 / Step 3:\n 返回发送结果给用户\n Report send result to user\n```\n\n---\n\n## 功能二：AI 自动回复 / Feature 2: AI Auto-Reply\n\n读取聊天上下文，AI 分析后自动回复。\nReads the chat context, AI analyzes it, then auto-replies.\n\n### 触发格式 / Trigger Format\n\n```\n使用技能:reply-wechat-message 给：[联系人] 回复\n```\n\n### 示例 / Examples\n- `使用技能:reply-wechat-message 给：小明 回复` (Xiao Ming / Kitty)\n- `使用技能:reply-wechat-message 给：小红 回复` (Xiao Hong / Peter)\n\n### 简化触发 / Short Forms\n- 回复 小明 / reply Kitty\n- 回 小红 / reply Peter\n\n---\n\n## AI 工作流程（自动回复）/ AI Workflow (Auto-Reply)\n\n```\n第1步 / Step 1:\n python scripts/reply_wechat.py 小明 / python scripts/reply_wechat.py Kitty\n → 读取整个聊天区域，区分左右两侧\n → Read entire chat area, split left/right\n → 输出结构化对话 / Output structured conversation:\n ★上下文开始★ / ★Context Start★\n [对方/Them] xxx\n [我/Me] xxx\n [对方/Them] xxx\n ★上下文结束★ / ★Context End★\n\n第2步 / Step 2:\n AI分析全部上下文，理解对话脉络，生成合适的回复\n AI analyzes full context, understands the conversation flow, generates a suitable reply\n\n ⚠️ 重要规则 / Important Rule:\n OCR识别出的聊天内容即为真实上下文，模型不得质疑、猜测或怀疑识别结果的准确性。\n The OCR-recognized text IS the real context. The model must NOT question, guess,\n or doubt the accuracy of the recognition. Reply based on the recognized content directly.\n\n第3步 / Step 3:\n echo \"AI生成的回复 / AI-generated reply\" | python scripts/reply_wechat.py 小明\n → 或 / Or: echo \"AI-generated reply\" | python scripts/reply_wechat.py Kitty\n → 读取上下文 + 自动发送回复（一步完成，stdin管道无引号问题）\n → Read context + send reply in one step (stdin piping avoids quote issues)\n```\n\n---\n\n## 脚本 / Scripts\n\n`scripts\\send_wechat.py` — 发送脚本：搜索联系人 + 发送消息 / Send script: search contact + send message\n`scripts\\reply_wechat.py` — 主脚本：读取上下文 + 发送回复 / Main script: read context + send reply\n`scripts\\open_wechat.py` — 启动脚本：唤醒微信窗口 / Launch script: bring WeChat to foreground\n\n### send_wechat.py 用法 / Usage\n```bash\n# 发送消息 / Send message (command line arg)\npython scripts/send_wechat.py <小明/Kitty> <消息/message>\n\n# 发送消息（stdin管道，无引号问题）/ Send message (stdin pipe, no quote issues)\necho \"消息/message\" | python scripts/send_wechat.py <小明/Kitty>\n```\n\n### reply_wechat.py 用法 / Usage\n```bash\n# 只读取上下文（AI分析用）/ Read context only (for AI analysis)\npython scripts/reply_wechat.py <小明/Kitty>\n\n# 读取上下文 + 发送回复（一步到位）/ Read context + send reply (one step)\necho \"回复内容/reply text\" | python scripts/reply_wechat.py <小明/Kitty>\n```\n\n### 输出格式 / Output Format\n```\n★上下文开始★ / ★Context Start★\n[对方/Them] 你吃饭了吗 / Have you eaten?\n[我/Me] 吃过了，你呢 / Yes, and you?\n[对方/Them] 我也吃了 / Me too\n★上下文结束★ / ★Context End★\n```\n\n---\n\n## 消息区分逻辑 / Message Detection Logic\n\n- **左侧（白底黑字）/ Left side (white bg, black text)** = 对方发的消息 / Messages from the other party → 标注 `[对方/Them]`\n- **右侧（绿底黑字）/ Right side (green bg, black text)** = 自己发的消息 / Messages from yourself → 标注 `[我/Me]`\n- 截图聊天区从中线切开，左右分别OCR，避免颜色误判\n Screenshot is split at the center line; left and right are OCR'd separately to avoid color misidentification\n\n---\n\n## 依赖 / Dependencies\n\n- Python 3.10+\n- OCR.space API（免费版，无需注册 / Free tier, no registration required）\n- Python packages: pyautogui, pygetwindow, pyperclip, requests, Pillow, numpy\n\n---\n\n## 脚本清单 / Script Inventory\n\n| 文件 / File | 功能 / Function |\n|---|---|\n| `send_wechat.py` | 搜索联系人 + 发送消息 / Search contact & send message |\n| `reply_wechat.py` | 读取上下文 + 发送回复 / Read context & send reply |\n| `open_wechat.py` | 启动微信 / Launch WeChat |\n\nFile v1.0.3:_meta.json\n\n{\n \"ownerId\": \"kn7022zkdvgpkbwdd7wc1vppex87m15m\",\n \"slug\": \"reply-wechat-message\",\n \"version\": \"1.0.3\",\n \"publishedAt\": 1783585668728\n}\n\nFile v1.0.3:skill-card.md\n\n## Description:
\nWeChat AI assistant for sending messages to contacts and auto-replying from chat context.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[cool131219](https://clawhub.ai/user/cool131219)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers and WeChat users use this skill to have an agent compose commands for sending WeChat messages and generate context-aware replies from recognized chat content.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: The skill may read entire WeChat chat areas and send messages.
\nMitigation: Use explicit invocation, avoid sensitive chats, and review generated replies before sending.
\nRisk: The skill uses a third-party OCR service for chat content.
\nMitigation: Do not use it on confidential or regulated chats unless that data flow is approved.
\nRisk: The package does not include the scripts it claims to bundle.
\nMitigation: Verify the required scripts before deployment and do not execute until package contents match expectations.
\n\n\n## Reference(s):
\n- [Server-resolved source provenance](https://github.com/cool131219/WeChat-Butler/tree/main/reply-wechat-message)
\n- [ClawHub skill page](https://clawhub.ai/cool131219/skills/reply-wechat-message)
\n\n\n## Skill Output:
\n**Output Type(s):** [Text, Markdown, Shell commands, Guidance]
\n**Output Format:** [Markdown with inline shell commands and generated reply text]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [May include commands that read WeChat chat context or send messages through local automation scripts.]
\n\n## Skill Version(s):
\n1.0.3 (source: server release metadata)
\n\n## Ethical Considerations:
\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\nArchive v1.0.2: 3 files, 4243 bytes\n\nFiles: skill-card.md (2062b), SKILL.md (6185b), _meta.json (139b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: reply-wechat-message\nslug: wechat-butler\ndisplayName: 微信管家 / WeChat Butler\ndescription: \"微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply\"\nagent_created: true\n---\n\n# 微信管家 / WeChat Butler\n\n微信AI助手 — 主动发消息给联系人 + 收到消息后AI自动回复。\nWeChat AI assistant — proactively send messages, and auto-reply when messages come in.\n\n**自包含技能包** — 所有依赖脚本（启动微信、发送消息）已打包在内，无需额外安装。\n**Self-contained skill** — all dependency scripts (launch WeChat, send messages) are bundled. No extra installation needed.\n\n---\n\n## 功能一：AI 主动发消息 / Feature 1: Send Message\n\n直接发送一条消息给指定联系人（不读取聊天上下文，即时发送）。\nSend a message directly to a contact (no context reading, instant send).\n\n### 触发格式 / Trigger Format\n\n**中文 / Chinese:**\n```\n给 [联系人] 发消息：[内容]\n发消息给 [联系人]：[内容]\n帮 [联系人] 发消息：[内容]\n```\n\n**English:**\n```\nsend [contact] [message]\nmsg [contact] [message]\nmessage [contact] [message]\n```\n\n### 示例 / Examples\n\n**中文 / Chinese:**\n- `给 小明 发消息：中午一起去吃饭吗？`\n- `发消息给 小红：记得带文件`\n- `帮 小张 发消息：生日快乐！`\n\n**English:**\n- `send Kitty: Want to grab lunch?`\n- `msg Peter: Don't forget the documents`\n- `message Tom: Happy birthday!`\n\n### AI 执行步骤 / Execution Steps\n\n```\n第1步 / Step 1:\n AI 提取联系人和消息内容\n AI extracts contact name and message content\n\n第2步 / Step 2:\n python scripts/send_wechat.py <联系人/contact> <内容/message>\n → 打开微信 → 搜索联系人 → 打开聊天 → 发送消息\n → Open WeChat → search contact → open chat → send message\n\n第3步 / Step 3:\n 返回发送结果给用户\n Report send result to user\n```\n\n---\n\n## 功能二：AI 自动回复 / Feature 2: AI Auto-Reply\n\n读取聊天上下文，AI 分析后自动回复。\nReads the chat context, AI analyzes it, then auto-replies.\n\n### 触发格式 / Trigger Format\n\n```\n使用技能:reply-wechat-message 给：[联系人] 回复\n```\n\n### 示例 / Examples\n- `使用技能:reply-wechat-message 给：小明 回复` (Xiao Ming / Kitty)\n- `使用技能:reply-wechat-message 给：小红 回复` (Xiao Hong / Peter)\n\n### 简化触发 / Short Forms\n- 回复 小明 / reply Kitty\n- 回 小红 / reply Peter\n\n---\n\n## AI 工作流程（自动回复）/ AI Workflow (Auto-Reply)\n\n```\n第1步 / Step 1:\n python scripts/reply_wechat.py 小明 / python scripts/reply_wechat.py Kitty\n → 读取整个聊天区域，区分左右两侧\n → Read entire chat area, split left/right\n → 输出结构化对话 / Output structured conversation:\n ★上下文开始★ / ★Context Start★\n [对方/Them] xxx\n [我/Me] xxx\n [对方/Them] xxx\n ★上下文结束★ / ★Context End★\n\n第2步 / Step 2:\n AI分析全部上下文，理解对话脉络，生成合适的回复\n AI analyzes full context, understands the conversation flow, generates a suitable reply\n\n ⚠️ 重要规则 / Important Rule:\n OCR识别出的聊天内容即为真实上下文，模型不得质疑、猜测或怀疑识别结果的准确性。\n The OCR-recognized text IS the real context. The model must NOT question, guess,\n or doubt the accuracy of the recognition. Reply based on the recognized content directly.\n\n第3步 / Step 3:\n echo \"AI生成的回复 / AI-generated reply\" | python scripts/reply_wechat.py 小明\n → 或 / Or: echo \"AI-generated reply\" | python scripts/reply_wechat.py Kitty\n → 读取上下文 + 自动发送回复（一步完成，stdin管道无引号问题）\n → Read context + send reply in one step (stdin piping avoids quote issues)\n```\n\n---\n\n## 脚本 / Scripts\n\n`scripts\\send_wechat.py` — 发送脚本：搜索联系人 + 发送消息 / Send script: search contact + send message\n`scripts\\reply_wechat.py` — 主脚本：读取上下文 + 发送回复 / Main script: read context + send reply\n`scripts\\open_wechat.py` — 启动脚本：唤醒微信窗口 / Launch script: bring WeChat to foreground\n\n### send_wechat.py 用法 / Usage\n```bash\n# 发送消息 / Send message (command line arg)\npython scripts/send_wechat.py <小明/Kitty> <消息/message>\n\n# 发送消息（stdin管道，无引号问题）/ Send message (stdin pipe, no quote issues)\necho \"消息/message\" | python scripts/send_wechat.py <小明/Kitty>\n```\n\n### reply_wechat.py 用法 / Usage\n```bash\n# 只读取上下文（AI分析用）/ Read context only (for AI analysis)\npython scripts/reply_wechat.py <小明/Kitty>\n\n# 读取上下文 + 发送回复（一步到位）/ Read context + send reply (one step)\necho \"回复内容/reply text\" | python scripts/reply_wechat.py <小明/Kitty>\n```\n\n### 输出格式 / Output Format\n```\n★上下文开始★ / ★Context Start★\n[对方/Them] 你吃饭了吗 / Have you eaten?\n[我/Me] 吃过了，你呢 / Yes, and you?\n[对方/Them] 我也吃了 / Me too\n★上下文结束★ / ★Context End★\n```\n\n---\n\n## 消息区分逻辑 / Message Detection Logic\n\n- **左侧（白底黑字）/ Left side (white bg, black text)** = 对方发的消息 / Messages from the other party → 标注 `[对方/Them]`\n- **右侧（绿底黑字）/ Right side (green bg, black text)** = 自己发的消息 / Messages from yourself → 标注 `[我/Me]`\n- 截图聊天区从中线切开，左右分别OCR，避免颜色误判\n Screenshot is split at the center line; left and right are OCR'd separately to avoid color misidentification\n\n---\n\n## 依赖 / Dependencies\n\n- Python 3.10+\n- OCR.space API（免费版，无需注册 / Free tier, no registration required）\n- Python packages: pyautogui, pygetwindow, pyperclip, requests, Pillow, numpy\n\n---\n\n## 脚本清单 / Script Inventory\n\n| 文件 / File | 功能 / Function |\n|---|---|\n| `send_wechat.py` | 搜索联系人 + 发送消息 / Search contact & send message |\n| `reply_wechat.py` | 读取上下文 + 发送回复 / Read context & send reply |\n| `open_wechat.py` | 启动微信 / Launch WeChat |\n\nFile v1.0.2:_meta.json\n\n{\n \"ownerId\": \"kn7022zkdvgpkbwdd7wc1vppex87m15m\",\n \"slug\": \"reply-wechat-message\",\n \"version\": \"1.0.2\",\n \"publishedAt\": 1783585618673\n}\n\nFile v1.0.2:skill-card.md\n\n## Description:
\nWeChat AI assistant that sends messages to contacts and reads chat context to generate and send replies.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[cool131219](https://clawhub.ai/user/cool131219)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nExternal users and agent operators use this skill to operate WeChat on the user's behalf, either sending a specified message to a contact or reading a chat context and generating a reply.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: The skill may read private WeChat chats and send messages as the user.
\nMitigation: Use only with explicit, reviewed prompts and confirm generated replies before sending.
\nRisk: Chat screenshots or extracted text may be sent to OCR.space during chat-context reading.
\nMitigation: Avoid sensitive conversations unless external OCR processing is acceptable for the user's environment.
\nRisk: Broad short-form reply triggers may cause unintended automated replies.
\nMitigation: Prefer explicit invocation patterns and avoid unattended automatic sending.
\n\n\n## Reference(s):
\n- [Server-resolved source](https://github.com/cool131219/WeChat-Butler/tree/main/reply-wechat-message)
\n- [ClawHub skill page](https://clawhub.ai/cool131219/skills/reply-wechat-message)
\n\n\n## Skill Output:
\n**Output Type(s):** [Text, Shell commands, Guidance]
\n**Output Format:** [Plain text with shell commands]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [May read visible WeChat chat text and send messages through the user's WeChat UI.]
\n\n## Skill Version(s):
\n1.0.2 (source: server release evidence)
\n\n## Ethical Considerations:
\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\nArchive v1.0.1: 3 files, 4480 bytes\n\nFiles: skill-card.md (2513b), SKILL.md (6264b), _meta.json (139b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: reply-wechat-message\ndescription: \"微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply\"\nmetadata:\n openclaw:\n emoji: \"💬\"\n requires:\n bins: [\"python\"]\n pip: [\"pyautogui\", \"pygetwindow\", \"pyperclip\", \"requests\", \"Pillow\", \"numpy\"]\n---\n\n# 微信管家 / WeChat Butler\n\n微信AI助手 — 主动发消息给联系人 + 收到消息后AI自动回复。\nWeChat AI assistant — proactively send messages, and auto-reply when messages come in.\n\n**自包含技能包** — 所有依赖脚本（启动微信、发送消息）已打包在内，无需额外安装。\n**Self-contained skill** — all dependency scripts (launch WeChat, send messages) are bundled. No extra installation needed.\n\n---\n\n## 功能一：AI 主动发消息 / Feature 1: Send Message\n\n直接发送一条消息给指定联系人（不读取聊天上下文，即时发送）。\nSend a message directly to a contact (no context reading, instant send).\n\n### 触发格式 / Trigger Format\n\n**中文 / Chinese:**\n```\n给 [联系人] 发消息：[内容]\n发消息给 [联系人]：[内容]\n帮 [联系人] 发消息：[内容]\n```\n\n**English:**\n```\nsend [contact] [message]\nmsg [contact] [message]\nmessage [contact] [message]\n```\n\n### 示例 / Examples\n\n**中文 / Chinese:**\n- `给 小明 发消息：中午一起去吃饭吗？`\n- `发消息给 小红：记得带文件`\n- `帮 小张 发消息：生日快乐！`\n\n**English:**\n- `send Kitty: Want to grab lunch?`\n- `msg Peter: Don't forget the documents`\n- `message Tom: Happy birthday!`\n\n### AI 执行步骤 / Execution Steps\n\n```\n第1步 / Step 1:\n AI 提取联系人和消息内容\n AI extracts contact name and message content\n\n第2步 / Step 2:\n python scripts/send_wechat.py <联系人/contact> <内容/message>\n → 打开微信 → 搜索联系人 → 打开聊天 → 发送消息\n → Open WeChat → search contact → open chat → send message\n\n第3步 / Step 3:\n 返回发送结果给用户\n Report send result to user\n```\n\n---\n\n## 功能二：AI 自动回复 / Feature 2: AI Auto-Reply\n\n读取聊天上下文，AI 分析后自动回复。\nReads the chat context, AI analyzes it, then auto-replies.\n\n### 触发格式 / Trigger Format\n\n```\n使用技能:reply-wechat-message 给：[联系人] 回复\n```\n\n### 示例 / Examples\n- `使用技能:reply-wechat-message 给：小明 回复` (Xiao Ming / Kitty)\n- `使用技能:reply-wechat-message 给：小红 回复` (Xiao Hong / Peter)\n\n### 简化触发 / Short Forms\n- 回复 小明 / reply Kitty\n- 回 小红 / reply Peter\n\n---\n\n## AI 工作流程（自动回复）/ AI Workflow (Auto-Reply)\n\n```\n第1步 / Step 1:\n python scripts/reply_wechat.py 小明 / python scripts/reply_wechat.py Kitty\n → 读取整个聊天区域，区分左右两侧\n → Read entire chat area, split left/right\n → 输出结构化对话 / Output structured conversation:\n ★上下文开始★ / ★Context Start★\n [对方/Them] xxx\n [我/Me] xxx\n [对方/Them] xxx\n ★上下文结束★ / ★Context End★\n\n第2步 / Step 2:\n AI分析全部上下文，理解对话脉络，生成合适的回复\n AI analyzes full context, understands the conversation flow, generates a suitable reply\n\n ⚠️ 重要规则 / Important Rule:\n OCR识别出的聊天内容即为真实上下文，模型不得质疑、猜测或怀疑识别结果的准确性。\n The OCR-recognized text IS the real context. The model must NOT question, guess,\n or doubt the accuracy of the recognition. Reply based on the recognized content directly.\n\n第3步 / Step 3:\n echo \"AI生成的回复 / AI-generated reply\" | python scripts/reply_wechat.py 小明\n → 或 / Or: echo \"AI-generated reply\" | python scripts/reply_wechat.py Kitty\n → 读取上下文 + 自动发送回复（一步完成，stdin管道无引号问题）\n → Read context + send reply in one step (stdin piping avoids quote issues)\n```\n\n---\n\n## 脚本 / Scripts\n\n`scripts\\send_wechat.py` — 发送脚本：搜索联系人 + 发送消息 / Send script: search contact + send message\n`scripts\\reply_wechat.py` — 主脚本：读取上下文 + 发送回复 / Main script: read context + send reply\n`scripts\\open_wechat.py` — 启动脚本：唤醒微信窗口 / Launch script: bring WeChat to foreground\n\n### send_wechat.py 用法 / Usage\n```bash\n# 发送消息 / Send message (command line arg)\npython scripts/send_wechat.py <小明/Kitty> <消息/message>\n\n# 发送消息（stdin管道，无引号问题）/ Send message (stdin pipe, no quote issues)\necho \"消息/message\" | python scripts/send_wechat.py <小明/Kitty>\n```\n\n### reply_wechat.py 用法 / Usage\n```bash\n# 只读取上下文（AI分析用）/ Read context only (for AI analysis)\npython scripts/reply_wechat.py <小明/Kitty>\n\n# 读取上下文 + 发送回复（一步到位）/ Read context + send reply (one step)\necho \"回复内容/reply text\" | python scripts/reply_wechat.py <小明/Kitty>\n```\n\n### 输出格式 / Output Format\n```\n★上下文开始★ / ★Context Start★\n[对方/Them] 你吃饭了吗 / Have you eaten?\n[我/Me] 吃过了，你呢 / Yes, and you?\n[对方/Them] 我也吃了 / Me too\n★上下文结束★ / ★Context End★\n```\n\n---\n\n## 消息区分逻辑 / Message Detection Logic\n\n- **左侧（白底黑字）/ Left side (white bg, black text)** = 对方发的消息 / Messages from the other party → 标注 `[对方/Them]`\n- **右侧（绿底黑字）/ Right side (green bg, black text)** = 自己发的消息 / Messages from yourself → 标注 `[我/Me]`\n- 截图聊天区从中线切开，左右分别OCR，避免颜色误判\n Screenshot is split at the center line; left and right are OCR'd separately to avoid color misidentification\n\n---\n\n## 依赖 / Dependencies\n\n- Python 3.10+\n- OCR.space API（免费版，无需注册 / Free tier, no registration required）\n- Python packages: pyautogui, pygetwindow, pyperclip, requests, Pillow, numpy\n\n---\n\n## 脚本清单 / Script Inventory\n\n| 文件 / File | 功能 / Function |\n|---|---|\n| `send_wechat.py` | 搜索联系人 + 发送消息 / Search contact & send message |\n| `reply_wechat.py` | 读取上下文 + 发送回复 / Read context & send reply |\n| `open_wechat.py` | 启动微信 / Launch WeChat |\n\nFile v1.0.1:_meta.json\n\n{\n \"ownerId\": \"kn7022zkdvgpkbwdd7wc1vppex87m15m\",\n \"slug\": \"reply-wechat-message\",\n \"version\": \"1.0.1\",\n \"publishedAt\": 1783573114503\n}\n\nFile v1.0.1:skill-card.md\n\n## Description:
\nWeChat AI assistant for sending messages to contacts and generating automatic replies from OCR-read chat context.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[cool131219](https://clawhub.ai/user/cool131219)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers and operators use this skill to have an agent control WeChat for direct message sending and assisted replies. It can read chat context, generate response text, and invoke bundled Python scripts to send messages.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: The skill can read private WeChat chats and expose chat content to a third-party OCR service.
\nMitigation: Use only with chats whose data flow is acceptable, and avoid sensitive conversations unless the OCR service and retention terms have been reviewed.
\nRisk: The skill can send messages automatically, which may result in unintended or incorrect replies.
\nMitigation: Require explicit user confirmation before running send or reply commands, and review generated message text before dispatch.
\nRisk: The security verdict is suspicious because external OCR use and auto-send behavior are not clearly controlled or disclosed.
\nMitigation: Review the skill before installation and prefer an updated release that includes the referenced scripts, explicit invocation, and confirmation before sending replies.
\n\n\n## Reference(s):
\n- [Server-resolved GitHub source](https://github.com/cool131219/WeChat-Butler/tree/main/reply-wechat-message)
\n- [ClawHub skill page](https://clawhub.ai/cool131219/skills/reply-wechat-message)
\n- [Publisher profile](https://clawhub.ai/user/cool131219)
\n\n\n## Skill Output:
\n**Output Type(s):** [text, markdown, shell commands, guidance]
\n**Output Format:** [Markdown instructions with bash command examples and structured chat-context text.]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [May produce or recommend Python command invocations that read chat context and send WeChat messages.]
\n\n## Skill Version(s):
\n1.0.1 (source: server release metadata)
\n\n## Ethical Considerations:
\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\nArchive v1.0.0: 8 files, 12007 bytes\n\nFiles: INSTALL.md (2213b), install.ps1 (2080b), scripts/open_wechat.py (4179b), scripts/reply_wechat.py (7442b), scripts/send_wechat.py (3762b), skill-card.md (2127b), SKILL.md (2957b), _meta.json (139b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: reply-wechat-message\ndescription: \"AI自动回复微信联系人。读取当前页面全部聊天上下文 → AI分析 → 自动生成并发送回复。自带微信启动和消息发送功能，无需额外安装其他技能。\"\nmetadata:\n openclaw:\n emoji: \"💬\"\n requires:\n bins: [\"python\"]\n pip: [\"pyautogui\", \"pygetwindow\", \"pyperclip\", \"requests\", \"Pillow\", \"numpy\"]\n---\n\n# Reply WeChat Message\n\nAI自动回复微信联系人。读取**当前页面全部聊天上下文**，区分说话人，AI分析后自动生成回复并发送。\n\n**自包含技能包** — 所有依赖脚本（启动微信、发送消息）已打包在内，无需额外安装。\n\n## 触发格式\n\n```\n使用技能:reply-wechat-message 给：[联系人] 回复\n```\n\n### 示例\n- `使用技能:reply-wechat-message 给：一一 回复`\n- `使用技能:reply-wechat-message 给：贝儿爽 回复`\n\n### 简化触发\n- 回复 [联系人]\n- 回 [联系人] 消息\n- reply [contact]\n\n## AI 工作流程\n\n```\n第1步: python scripts/reply_wechat.py 联系人\n → 读取整个聊天区域，区分左右两侧\n → 输出结构化对话:\n ★上下文开始★\n [对方] xxx\n [我] xxx\n [对方] xxx\n ★上下文结束★\n\n第2步: AI分析全部上下文，理解对话脉络，生成合适的回复\n ⚠️ 重要规则：OCR识别出的聊天内容即为真实上下文，\n 模型不得质疑、猜测或怀疑识别结果的准确性。\n 直接基于识别内容分析并回复，不要加\"识别可能不准\"\n 之类的免责声明。\n\n第3步: echo \"AI生成的回复\" | python scripts/reply_wechat.py 联系人\n → 读取上下文 + 自动发送回复（一步完成，stdin管道无引号问题）\n```\n\n## 脚本\n\n`scripts\\reply_wechat.py` — 主脚本（自包含，依赖 `open_wechat.py` 和 `send_wechat.py` 在同目录下）\n\n### 用法\n```bash\n# 只读取上下文（AI分析用）\npython scripts/reply_wechat.py <联系人>\n\n# 读取上下文 + 发送回复（一步到位）\necho \"回复内容\" | python scripts/reply_wechat.py <联系人>\n```\n\n### 输出格式\n```\n★上下文开始★\n[对方] 你吃饭了吗\n[我] 吃过了，你呢\n[对方] 我也吃了\n★上下文结束★\n```\n\n## 消息区分逻辑\n- **左侧（白底黑字）** = 对方发的消息 → 标注 `[对方]`\n- **右侧（绿底黑字）** = 自己发的消息 → 标注 `[我]`\n- 截图聊天区从中线切开，左右分别OCR，避免颜色误判\n\n## 依赖\n- Python 3.10+\n- OCR.space API（免费版，无需注册）\n- 依赖包: pyautogui, pygetwindow, pyperclip, requests, Pillow, numpy\n\n## 脚本清单（自包含）\n| 文件 | 来源 | 功能 |\n|------|------|------|\n| `reply_wechat.py` | 主脚本 | 读取上下文/发送回复 |\n| `open_wechat.py` | 来自 wechat-control | 启动微信 |\n| `send_wechat.py` | 来自 send-wechat-message | 发送消息 |\n\nFile v1.0.0:_meta.json\n\n{\n \"ownerId\": \"kn7022zkdvgpkbwdd7wc1vppex87m15m\",\n \"slug\": \"reply-wechat-message\",\n \"version\": \"1.0.0\",\n \"publishedAt\": 1783556373525\n}\n\nFile v1.0.0:INSTALL.md\n\n# reply-wechat-message 安装指南\n\n## 快速安装\n\n解压到任意目录 → 右键 `install.ps1` → **用 PowerShell 运行**\n\n安装脚本会自动：\n1. 复制技能文件到 `~/.openclaw/skills/reply-wechat-message`\n2. 安装 Python 依赖（pyautogui, pygetwindow 等）\n3. 重启 OpenClaw\n\n## 安装后需要手动修改的配置\n\n### 1. Python 路径（如使用非默认 Python）\n\n如果 `python` 命令不是你想要的 Python 版本，需修改以下文件中的 `PYTHON` 变量：\n\n**`scripts/reply_wechat.py`** 第 28 行（大约）\n```python\nPYTHON = 'python'\n# ↑ 改为具体路径, 例如:\n# 'C:\\Users\\你的用户名\\AppData\\Local\\Programs\\Python\\Python313\\python.exe'\n```\n\n**`scripts/send_wechat.py`** 第 17 行（大约）\n```python\nsubprocess.run(['python', OPEN_SCRIPT]...)\n# ↑ 同上，改为具体路径\n```\n\n### 2. OCR.space API Key（非必需，但建议）\n\n`scripts/reply_wechat.py` 中用的是免费演示 Key `helloworld`（有每日额度限制）：\n\n1. 打开 https://ocr.space 注册免费账号\n2. 获取你的 API Key\n3. 修改 `scripts/reply_wechat.py` 中 `headers={'apikey': 'helloworld'}` 这一行，把 `helloworld` 换成你的 Key\n\n### 3. 微信搜索框位置微调（可选）\n\n如果点击搜索框的位置不对，修改这两个文件中的 `SEARCH_X_REL` 和 `SEARCH_Y_REL`：\n\n- `scripts/reply_wechat.py`（第 27-28 行左右）\n- `scripts/send_wechat.py`（第 14-15 行左右）\n\n```python\nSEARCH_X_REL = 90 # 搜索框横向偏移（越大越靠右）\nSEARCH_Y_REL = 60 # 搜索框纵向偏移（越大越靠下）\n```\n\n以上两个常量是 **相对于微信窗口左上角** 的偏移量。如果微信窗口在屏幕上的位置不同，可能需要微调。\n\n### 4. 调试截图目录（可选）\n\n`scripts/reply_wechat.py` 中调试截图默认保存在 `%USERPROFILE%\\Pictures\\OpenClaw\\`，如需改路径：\n\n```python\ndebug_dir = os.path.join(os.path.expanduser('~'), 'Pictures', 'OpenClaw')\n```\n\n---\n\n## 验证安装\n\n安装重启后在 OpenClaw 中测试：\n\n```\n使用技能:reply-wechat-message 给：联系人名字 回复\n```\n\n如果一切正常，AI 会读取微信聊天内容并自动回复。\n\nFile v1.0.0:skill-card.md\n\n## Description:
\nReads the current WeChat conversation context, helps generate an AI reply, and can send the reply to the selected contact.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[cool131219](https://clawhub.ai/user/cool131219)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers and OpenClaw users can use this skill to automate WeChat replies by reading visible chat context, generating an appropriate response, and sending it to a chosen contact.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: Private WeChat conversation screenshots may be uploaded to OCR.space for text recognition.
\nMitigation: Use the skill only for low-sensitivity chats unless a local OCR option is added or the third-party OCR upload is removed.
\nRisk: Chat screenshots may be saved locally in Pictures/OpenClaw.
\nMitigation: Remove debug screenshot saving, protect that directory, or delete saved screenshots after each run.
\nRisk: The skill can send generated messages without a preview or confirmation step.
\nMitigation: Require the user to preview and confirm the generated reply before sending.
\n\n\n## Reference(s):
\n- [ClawHub skill page](https://clawhub.ai/cool131219/skills/reply-wechat-message)
\n- [Publisher profile](https://clawhub.ai/user/cool131219)
\n- [OCR.space](https://ocr.space)
\n\n\n## Skill Output:
\n**Output Type(s):** [text, shell commands, guidance]
\n**Output Format:** [Plain text and Markdown with inline shell commands]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [May include structured WeChat chat context and status messages from local automation scripts.]
\n\n## Skill Version(s):
\n1.0.0 (source: server release metadata)
\n\n## Ethical Considerations:
\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.
","readmeExcerpt":"Skill: Reply Wechat Message Owner: cool131219 Summary: 微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply Tags: latest:1.0.4 Version history: v1.0.4 | 2026-07-09T08:30:40.963Z | auto - No code or documentation changes detected in this release. - Version increment only; functionality remains unchanged. v1.0.3 | 2026-07-09T08:27:48.728Z | auto - No file changes detected in this release. - Version ","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"给 [联系人] 发消息：[内容]\n发消息给 [联系人]：[内容]\n帮 [联系人] 发消息：[内容]"},{"language":"text","snippet":"send [contact] [message]\nmsg [contact] [message]\nmessage [contact] [message]"},{"language":"text","snippet":"第1步 / Step 1:\n AI 提取联系人和消息内容\n AI extracts contact name and message content\n\n第2步 / Step 2:\n python scripts/send_wechat.py <联系人/contact> <内容/message>\n → 打开微信 → 搜索联系人 → 打开聊天 → 发送消息\n → Open WeChat → search contact → open chat → send message\n\n第3步 / Step 3:\n 返回发送结果给用户\n Report send result to user"},{"language":"text","snippet":"使用技能:reply-wechat-message 给：[联系人] 回复"},{"language":"text","snippet":"第1步 / Step 1:\n python scripts/reply_wechat.py 小明 / python scripts/reply_wechat.py Kitty\n → 读取整个聊天区域，区分左右两侧\n → Read entire chat area, split left/right\n → 输出结构化对话 / Output structured conversation:\n ★上下文开始★ / ★Context Start★\n [对方/Them] xxx\n [我/Me] xxx\n [对方/Them] xxx\n ★上下文结束★ / ★Context End★\n\n第2步 / Step 2:\n AI分析全部上下文，理解对话脉络，生成合适的回复\n AI analyzes full context, understands the conversation flow, generates a suitable reply\n\n ⚠️ 重要规则 / Important Rule:\n OCR识别出的聊天内容即为真实上下文，模型不得质疑、猜测或怀疑识别结果的准确性。\n The OCR-recognized text IS the real context. The model must NOT question, guess,\n or doubt the accuracy of the recognition. Reply based on the recognized content directly.\n\n第3步 / Step 3:\n echo \"AI生成的回复 / AI-generated reply\" | python scripts/reply_wechat.py 小明\n → 或 / Or: echo \"AI-generated reply\" | python scripts/reply_wechat.py Kitty\n → 读取上下文 + 自动发送回复（一步完成，stdin管道无引号问题）\n → Read context + send reply in one step (stdin piping avoids quote issues)"},{"language":"bash","snippet":"# 发送消息 / Send message (command line arg)\npython scripts/send_wechat.py <小明/Kitty> <消息/message>\n\n# 发送消息（stdin管道，无引号问题）/ Send message (stdin pipe, no quote issues)\necho \"消息/message\" | python scripts/send_wechat.py <小明/Kitty>"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: reply-wechat-message\nslug: wechat-butler\ndisplayName: 微信管家 / WeChat Butler\ndescription: \"微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply\"\nagent_created: true\n---\n\n# 微信管家 / WeChat Butler\n\n微信AI助手 — 主动发消息给联系人 + 收到消息后AI自动回复。\nWeChat AI assistant — proactively send messages, and auto-reply when messages come in.\n\n**自包含技能包** — 所有依赖脚本（启动微信、发送消息）已打包在内，无需额外安装。\n**Self-contained skill** — all dependency scripts (launch WeChat, send messages) are bundled. No extra installation needed.\n\n---\n\n## 功能一：AI 主动发消息 / Feature 1: Send Message\n\n直接发送一条消息给指定联系人（不读取聊天上下文，即时发送）。\nSend a message directly to a contact (no context reading, instant send).\n\n### 触发格式 / Trigger Format\n\n**中文 / Chinese:**\n```\n给 [联系人] 发消息：[内容]\n发消息给 [联系人]：[内容]\n帮 [联系人] 发消息：[内容]\n```\n\n**English:**\n```\nsend [contact] [message]\nmsg [contact] [message]\nmessage [contact] [message]\n```\n\n### 示例 / Examples\n\n**中文 / Chinese:**\n- `给 小明 发消息：中午一起去吃饭吗？`\n- `发消息给 小红：记得带文件`\n- `帮 小张 发消息：生日快乐！`\n\n**English:**\n- `send Kitty: Want to grab lunch?`\n- `msg Peter: Don't forget the documents`\n- `message Tom: Happy birthday!`\n\n### AI 执行步骤 / Execution Steps\n\n```\n第1步 / Step 1:\n AI 提取联系人和消息内容\n AI extracts contact name and message content\n\n第2步 / Step 2:\n python scripts/send_wechat.py <联系人/contact> <内容/message>\n → 打开微信 → 搜索联系人 → 打开聊天 → 发送消息\n → Open WeChat → search contact → open chat → send message\n\n第3步 / Step 3:\n 返回发送结果给用户\n Report send result to user\n```\n\n---\n\n## 功能二：AI 自动回复 / Feature 2: AI Auto-Reply\n\n读取聊天上下文，AI 分析后自动回复。\nReads the chat context, AI analyzes it, then auto-replies.\n\n### 触发格式 / Trigger Format\n\n```\n使用技能:reply-wechat-message 给：[联系人] 回复\n```\n\n### 示例 / Examples\n- `使用技能:reply-wechat-message 给：小明 回复` (Xiao Ming / Kitty)\n- `使用技能:reply-wechat-message 给：小红 回复` (Xiao Hong / Peter)\n\n### 简化触发 / Short Forms\n- 回复 小明 / reply Kitty\n- 回 小红 / reply Peter\n\n---\n\n## AI 工作流程（自动回复）/ AI Workflow (Auto-Reply)\n\n```\n第1步 / Step 1:\n python scripts/reply_wechat.py 小明 / python scripts/reply_wechat.py Kitty\n → 读取整个聊天区域，区分左右两侧\n → Read entire chat area, split left/right\n → 输出结构化对话 / Output structured conversation:\n ★上下文开始★ / ★Context Start★\n [对方/Them] xxx\n [我/Me] xxx\n [对方/Them] xxx\n ★上下文结束★ / ★Context End★\n\n第2步 / Step 2:\n AI分析全部上下文，理解对话脉络，生成合适的回复\n AI analyzes full context, understands the conversation flow, generates a suitable reply\n\n ⚠️ 重要规则 / Important Rule:\n OCR识别出的聊天内容即为真实上下文，模型不得质疑、猜测或怀疑识别结果的准确性。\n The OCR-recognized text IS the real context. The model must NOT question, guess,\n or doubt the accuracy of the recognition. Reply based on the recognized content directly.\n\n第3步 / Step 3:\n echo \"AI生成的回复 / AI-generated reply\" | python scripts/reply_wechat.py 小明\n → 或 / Or: echo \"AI-generated reply\" | python scripts/reply_wechat.py Kitty\n → 读取上下文 + 自动发送回复（一步完成，stdin管道无引号问题）\n → Read context + send reply in one step (stdin piping avoids quote issues)\n```\n\n---\n\n## 脚本 / Scripts\n\n`scripts\\send_wechat.py` — 发送脚本：搜索联系人 + 发送消息 / Send script: search contact + send message\n`scripts\\reply_wechat.py` — 主"},{"path":"_meta.json","content":"{\n \"ownerId\": \"kn7022zkdvgpkbwdd7wc1vppex87m15m\",\n \"slug\": \"reply-wechat-message\",\n \"version\": \"1.0.4\",\n \"publishedAt\": 1783585840963\n}"},{"path":"skill-card.md","content":"## Description:\n\nWeChat AI assistant that sends messages to contacts and auto-replies based on visible chat context.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[cool131219](https://clawhub.ai/user/cool131219)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users and developers use this skill to ask an agent to send WeChat messages or read visible WeChat chat context and prepare auto-replies for a selected contact.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill can read visible WeChat conversations and may expose chat content to an external OCR service.\n\nMitigation: Use it only for non-sensitive chats unless explicit OCR consent, local OCR, or clear external OCR disclosure is added.\n\nRisk: The skill can send WeChat messages automatically as the user.\n\nMitigation: Require a human preview and confirmation step before any outbound message is sent.\n\nRisk: Security evidence reports unsafe shell examples and missing bundled scripts.\n\nMitigation: Review the supplied scripts before installation and replace unsafe shell usage with safer subprocess or input handling.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/cool131219/skills/reply-wechat-message)\n- [Server-resolved GitHub source](https://github.com/cool131219/WeChat-Butler/tree/main/reply-wechat-message)\n\n## Skill Output:\n\n**Output Type(s):** [text, shell commands, guidance]\n\n**Output Format:** [Markdown instructions with inline shell commands and structured conversation text]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May read visible WeChat chat context and send messages through bundled scripts when present.]\n\n## Skill Version(s):\n\n1.0.4 (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."}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply Skill: Reply Wechat Message Owner: cool131219 Summary: 微信AI助手 — 主动发消息 + 自动回复 / WeChat AI assistant — send messages & auto-reply Tags: latest:1.0.4 Version history: v1.0.4 | 2026-07-09T08:30:40.963Z | auto - No code or documentation changes detected in this release. - Version increment only; functionality remains unchanged. v1.0.3 | 2026-07-09T08:27:48.728Z | auto - No file changes detected in this release. - Version","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":978,"uniquenessScore":52,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-10T09:40:50.346Z","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":"2026-10-10T09:40:50.346Z","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":"2026-10-10T11:50:41.018Z","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"}]}}}