{"id":"b7c84c26-d560-41b1-a6f5-b959cc8caa5d","entityType":"agent","slug":"clawhub-gecho-ai-tiktok-search","name":"tiktok-search","canonicalUrl":"https://www.xpersona.co/agent/clawhub-gecho-ai-tiktok-search","canonicalPath":"/agent/clawhub-gecho-ai-tiktok-search","generatedAt":"2026-10-10T08:47:29.021Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-10T04:10:57.267Z","emptyReason":null},"description":"Search TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server.","descriptionLabel":"Source description","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.7K downloads reported by the source. Last updated 10/10/2026.","installCommand":"clawhub skill install s175fhc7pkv1bdmpprb4xxy1ch86dc5q:tiktok-search","sourceUrl":"https://clawhub.ai/gecho-ai/tiktok-search","homepage":"https://clawhub.ai/gecho-ai/skills/tiktok-search","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/gecho-ai/tiktok-search","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/gecho-ai/skills/tiktok-search","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":65,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"tiktok-search technical dossier on Xpersona with agent coverage, OPENCLEW support, and live trust metadata."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-10T04:10:57.267Z","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":"2026-10-10T04:10:57.267Z","emptyReason":null},"stars":null,"forks":null,"downloads":1699,"packageName":null,"latestVersion":"1.1.38","tractionLabel":"1.7K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-10T04:10:57.266Z","emptyReason":null},"lastUpdatedAt":"2026-10-10T04:10:57.267Z","lastCrawledAt":"2026-10-10T04:10:57.266Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-11T04:10:57.266Z","lastVerifiedAt":null,"highlights":[{"version":"1.1.38","createdAt":"2026-09-23T08:58:48.634Z","changelog":"Release 1.1.38","fileCount":3,"zipByteSize":8782},{"version":"1.1.37","createdAt":"2026-08-21T15:05:02.837Z","changelog":"- Added new capability to retrieve detail data, comments, and replies for a single TikTok video via the `tiktok_video` tool. - Updated documentation to include guidance and prompts for collecting one-video detail and comment data. - Best-fit prompts now reference the new feature for retrieving video details by URL. - Removed legacy file: `skill-card.md`. - Minor clarifications and tool documentation updates for accuracy and coverage.","fileCount":3,"zipByteSize":8877},{"version":"1.1.36","createdAt":"2026-08-16T05:09:05.902Z","changelog":"Release 1.1.36","fileCount":3,"zipByteSize":7899},{"version":"1.1.30","createdAt":"2026-08-01T08:10:51.775Z","changelog":"Release 1.1.30 team owner verification","fileCount":3,"zipByteSize":8235},{"version":"1.1.29","createdAt":"2026-07-21T09:28:38.812Z","changelog":"- Added support for collecting videos from a specific TikTok creator via the new `tiktok_influencer` tool. - Updated documentation to cover creator video collection and tool usage. - Expanded official support/resource links, including an additional 1:1 support QR code. - Removed the sample skill card file for simpler distribution.","fileCount":3,"zipByteSize":7937},{"version":"1.1.24","createdAt":"2026-06-06T14:28:29.312Z","changelog":"**Changelog for tiktok-search v1.1.24** - The setup section was rewritten for greater clarity and simplicity, emphasizing the three critical steps to enable the Skill (install Gecho Chrome extension, log in to Gecho extension, log in to TikTok). - More explicit warnings and instructions were added about required extension and account logins for proper operation. - Minor language and formatting improvements were made for better readability. - The file skill-card.md was removed.","fileCount":3,"zipByteSize":7746},{"version":"1.1.23","createdAt":"2026-06-06T03:43:49.251Z","changelog":"tiktok-search v1.1.23 - Removed the skill-card.md file to streamline documentation. - SKILL.md rewritten and condensed: install/setup instructions moved up, redundant sections removed, and language clarified for faster onboarding. - Guidance for OpenClaw and Hermes setup made more concise; only official Gecho MCP methods are described. - Core workflow unchanged: search, insight, and status tools remain the supported entry points. - Official support links and resources are now collected at the top for easier access.","fileCount":3,"zipByteSize":6676},{"version":"1.1.22","createdAt":"2026-05-25T09:10:31.177Z","changelog":"tiktok-search 1.1.22 changelog: - Clarified that the OpenClaw plugin installation method is optional, not recommended as the primary setup path. - Updated setup instructions to emphasize the primary MCP configuration commands (`openclaw mcp set` for OpenClaw, `hermes mcp add` for Hermes). - Improved agent guidance to mention the plugin route only when users explicitly request plugin-based management. - No code or feature changes; instructional/documentation improvements only.","fileCount":3,"zipByteSize":8245}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s175fhc7pkv1bdmpprb4xxy1ch86dc5q:tiktok-search","setupComplexity":"low","setupSteps":["Install using `clawhub skill install s175fhc7pkv1bdmpprb4xxy1ch86dc5q:tiktok-search` 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/gecho-ai/tiktok-search 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-gecho-ai-tiktok-search/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T08:47:29.016Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gecho-ai-tiktok-search/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":"2026-10-10T04:10:57.267Z","emptyReason":null},"readme":"Skill: tiktok-search\n\nOwner: gecho-ai\n\nSummary: Search TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server.\n\nTags: latest:1.1.38\n\nVersion history:\n\nv1.1.38 | 2026-09-23T08:58:48.634Z | user\n\nRelease 1.1.38\n\nv1.1.37 | 2026-08-21T15:05:02.837Z | auto\n\n- Added new capability to retrieve detail data, comments, and replies for a single TikTok video via the `tiktok_video` tool.\n- Updated documentation to include guidance and prompts for collecting one-video detail and comment data.\n- Best-fit prompts now reference the new feature for retrieving video details by URL.\n- Removed legacy file: `skill-card.md`.\n- Minor clarifications and tool documentation updates for accuracy and coverage.\n\nv1.1.36 | 2026-08-16T05:09:05.902Z | user\n\nRelease 1.1.36\n\nv1.1.30 | 2026-08-01T08:10:51.775Z | user\n\nRelease 1.1.30 team owner verification\n\nv1.1.29 | 2026-07-21T09:28:38.812Z | user\n\n- Added support for collecting videos from a specific TikTok creator via the new `tiktok_influencer` tool.\n- Updated documentation to cover creator video collection and tool usage.\n- Expanded official support/resource links, including an additional 1:1 support QR code.\n- Removed the sample skill card file for simpler distribution.\n\nv1.1.24 | 2026-06-06T14:28:29.312Z | user\n\n**Changelog for tiktok-search v1.1.24**\n\n- The setup section was rewritten for greater clarity and simplicity, emphasizing the three critical steps to enable the Skill (install Gecho Chrome extension, log in to Gecho extension, log in to TikTok).\n- More explicit warnings and instructions were added about required extension and account logins for proper operation.\n- Minor language and formatting improvements were made for better readability.\n- The file skill-card.md was removed.\n\nv1.1.23 | 2026-06-06T03:43:49.251Z | user\n\ntiktok-search v1.1.23\n\n- Removed the skill-card.md file to streamline documentation.\n- SKILL.md rewritten and condensed: install/setup instructions moved up, redundant sections removed, and language clarified for faster onboarding.\n- Guidance for OpenClaw and Hermes setup made more concise; only official Gecho MCP methods are described.\n- Core workflow unchanged: search, insight, and status tools remain the supported entry points.\n- Official support links and resources are now collected at the top for easier access.\n\nv1.1.22 | 2026-05-25T09:10:31.177Z | user\n\ntiktok-search 1.1.22 changelog:\n\n- Clarified that the OpenClaw plugin installation method is optional, not recommended as the primary setup path.\n- Updated setup instructions to emphasize the primary MCP configuration commands (`openclaw mcp set` for OpenClaw, `hermes mcp add` for Hermes).\n- Improved agent guidance to mention the plugin route only when users explicitly request plugin-based management.\n- No code or feature changes; instructional/documentation improvements only.\n\nv1.1.21 | 2026-05-22T10:49:29.083Z | user\n\nSummary: This update adds stricter setup enforcement, multi-platform (OpenClaw/Hermes) support, and explicit invocation requirements.\n\n- Enforces an absolute trigger constraint: this skill must be invoked before calling any Gecho MCP tools for TikTok-related tasks.\n- Adds setup instructions and command examples for both OpenClaw and Hermes agent environments.\n- Requires agents to check readiness and stop with guidance if the official Gecho MCP tools are not available, disallowing fallback to terminal, browser, or web scrapers.\n- Clarifies guidance to never improvise installation or verification steps; only specific commands from the instructions may be provided.\n- Improves setup awareness and troubleshooting instructions for ClawHub, OpenClaw, and Hermes scenarios.\n\nv1.1.19 | 2026-05-13T08:02:33.723Z | user\n\n**Summary:** Major update focused on improving user guidance and clarifying prerequisites. Official installation guide now external.  \n\n- Consolidated user and agent setup steps, moving most technical instructions and troubleshooting to a linked README.\n- Added clear step-by-step onboarding with commands for OpenClaw-based deployments.\n- Improved error handling and first-use instructions for both direct users and agents.\n- Introduced a new section for most common use cases and recommended starter prompts.\n- Added explicit mention of `check_insight_status` tool for async job retrieval.\n- Provided official Discord and WeChat links for real-time community support.\n\nv1.1.7 | 2026-05-09T10:29:34.116Z | user\n\nInitial release of the TikTok Search skill – professional TikTok keyword search and extraction tool.\n\n- Supports automated TikTok search and metadata extraction through a Chrome extension and Node.js.\n- Provides two main tools: `tiktok_search` (video metadata) and `tiktok_insight` (trend/business analysis).\n- Enforces strict execution rules: one tool call per turn, no retries, no fallbacks, and robust error handling with explicit reporting.\n- Returns concise summaries in chat; saves full results to a specified local path.\n- Includes step-by-step usage instructions, error troubleshooting guidelines, and highlights all operational constraints.\n\nArchive index:\n\nArchive v1.1.38: 3 files, 8782 bytes\n\nFiles: _meta.json (133b), skill-card.md (2658b), SKILL.md (22727b)\n\nFile v1.1.38:SKILL.md\n\n---\nname: tiktok-search\ndescription: Search TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server.\nmetadata:\n  openclaw:\n    os: [\"darwin\", \"linux\", \"win32\"]\n    requires:\n      bins: [\"node\", \"npx\"]\n  hermes:\n    tags: [tiktok, search, insight, gecho, mcp]\n    category: social-media\n    os: [darwin, linux, windows]\n---\n\n# TikTok Search & Insight by Gecho\n\nSearch TikTok from an AI chat, collect structured video metadata, and run async product or trend insight jobs through the official Gecho Bridge MCP workflow.\n\nThis is the default TikTok aggregate Skill for Gecho. It covers TikTok video search, creator collection, one-video detail and comment collection, and TikTok insight. Single-tool TikTok Skills may exist for distribution and search traffic, but this Skill is the recommended default for users who want the complete TikTok research workflow.\n\n## Critical prerequisite: read before use\n\nGecho Skills must be used with the Gecho Chrome extension. You must be logged in to both your Gecho account in the extension and TikTok in the Chrome web app. If either login is missing, all TikTok search and insight features may fail even though the Skill is installed.\n\n## 3-step quick start\n\n### Step 1: Install the Gecho Chrome extension\n\n1. Open the [Gecho Chrome extension download page](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb).\n2. Click `Add to Chrome`, then confirm `Add extension`.\n\n### Step 2: Log in to the Gecho extension\n\nOpen the Gecho extension in Chrome and log in to your Gecho account. Keep the extension online.\n\n### Step 3: Log in to TikTok web\n\nOpen TikTok in Chrome and log in to the TikTok web app. Keep the logged-in TikTok tab open.\n\nAfter setup is complete, return to the OpenClaw dashboard or Hermes and ask: \"Search xxx on TikTok\". For example: \"Search hamburgers on TikTok\".\n\n## Official links and setup help\n\n- Website: [gecho.ai](https://gecho.ai/)\n- GitHub: [gecho-ai/gecho-bridge](https://github.com/gecho-ai/gecho-bridge)\n- YouTube channel: [@Gecho-AI](https://www.youtube.com/@Gecho-AI)\n- Chrome extension: [Gecho Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- OpenClaw setup video: [OpenClaw + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- Hermes setup video: [Hermes + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n- Discord: [https://discord.gg/RFDVZMR6Tn](https://discord.gg/RFDVZMR6Tn)\n- WeCom group QR code: [qywx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg)\n- 1:1 support QR code: [wx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg)\n\n## What this skill does\n\n- Finds high-performing TikTok videos for a keyword.\n- Collects titles, authors, engagement data, and video links.\n- Retrieves detail data, comments, and replies for one known TikTok video URL.\n- Saves full raw result sets to a local JSON file.\n- Starts async insight jobs for product research, competitor analysis, and trend discovery.\n\nBest-fit prompts:\n\n- \"Search TikTok for portable blender and show the top liked videos.\"\n- \"Find winning hooks for cat toy videos.\"\n- \"Run product opportunity insight for outdoor picnic mat.\"\n- \"Check the status of my previous TikTok insight job.\"\n- \"Get the details and comments for this TikTok video: https://www.tiktok.com/@user/video/123.\"\n\n## Important: Skill-only install is not enough\n\nThis Skill is the instruction layer. It tells the AI when and how to use Gecho.\n\nTo actually run TikTok searches, video-detail collection, or insight jobs, the user also needs:\n\n- the Gecho Bridge MCP server\n- the Gecho Chrome extension\n- Chrome with TikTok logged in\n- the Gecho extension logged in to a Gecho account and online\n\nIf the user installed only this Skill from ClawHub, searches will not work until the Gecho Bridge MCP server is configured. In that case, use the MCP setup path below.\n\nAlready installed Gecho Bridge? If `@gecho-ai/gecho-bridge-bundle` is installed and the Gecho MCP tools are visible, no extra MCP setup is needed for this Skill.\n\n## Quick start\n\n### OpenClaw Skill install: configure MCP\n\nIf this Skill is already installed in OpenClaw, configure the Gecho Bridge MCP server once:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\n### Optional: OpenClaw Bundle Plugin\n\nIf the user has not installed this Skill yet and prefers plugin management, the bundle plugin can install Gecho with the MCP entry configured:\n\n```bash\nopenclaw plugins install clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\nTo upgrade later:\n\n```bash\nopenclaw plugins update clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\n### Hermes setup\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\nIf Hermes cannot find `npx` even though Node.js is installed, use the absolute `npx` path. On many macOS Homebrew installs this is:\n\n```bash\nhermes mcp add gecho-bridge --command /opt/homebrew/bin/npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n## First-run checklist\n\nBefore the first search, make sure:\n\n- Node.js `>= 18` is available.\n- The [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) is installed.\n- Chrome is open with TikTok logged in.\n- The Gecho extension is logged in to a Gecho account and online.\n- The TikTok tab is not blocked by CAPTCHA, login walls, or a frozen page.\n\nFull setup guide:\n[Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n\nFor videos and support links, see the official links section above.\n\n## Tool choice\n\n| User goal | Use tool | Notes |\n|---|---|---|\n| Search TikTok videos and collect metadata | `tiktok_search` | Returns results directly |\n| Collect detail data, comments, and replies from one known video | `tiktok_video` | Requires a TikTok video URL |\n| Collect videos from a TikTok creator | `tiktok_influencer` | Uses the creator's `uniqueId` |\n| Analyze a niche, product, trend, or competitor opportunity | `tiktok_insight` | Starts an async job and returns a `jobId` |\n| Check an existing async insight job | `check_insight_status` | Use the `jobId` returned by `tiktok_insight` |\n\nUse this aggregate Skill for broad TikTok research requests. If the user explicitly asks for one exact raw tool only, such as \"run tiktok_video\", a matching single-tool distribution Skill may be used instead when available.\n\n## Official MCP tools\n\n### `tiktok_search`\n\nSearches TikTok for a keyword, scrolls the page through the Gecho browser extension, returns structured metadata, and saves the full result set.\n\nParameters:\n\n- `query` string, required: search keyword or phrase.\n- `targetCount` number, optional: requested result count; defaults to `100`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A JSON array of video metadata, plus a saved local file path when results are written successfully.\n\n### `tiktok_video`\n\nGets TikTok video detail data and comments. It opens the video detail page, expands comments, and continues scrolling to collect comments and replies.\n\nParameters:\n\n- `url` string, required: TikTok video detail-page URL, for example `https://www.tiktok.com/@user/video/123...`.\n- `targetCount` number, optional: maximum number of comments and replies to collect; default and maximum are `200`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- Structured video detail data, comments, and replies, plus a saved local file path when results are written successfully.\n\n### `tiktok_influencer`\n\nCollects videos published by one TikTok creator.\n\nParameters:\n\n- `uniqueId` string, required: the creator's TikTok `uniqueId` (for example, `zachking`).\n- `targetCount` number, optional: requested video count; defaults to `100`. Keep requests at or below `500`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\n### `tiktok_insight`\n\nStarts an asynchronous TikTok insight job for product, trend, or competitor research.\n\nParameters:\n\n- `query` string, required: search keyword or product/category phrase.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A `jobId`. The final result must be checked later with `check_insight_status`.\n\n### `check_insight_status`\n\nChecks the status or final result of an existing insight job.\n\nParameters:\n\n- `jobId` string, required: the job ID returned by `tiktok_insight`.\n\nExpected result:\n\n- `running`, `error`, or completed insight data.\n\n## Agent execution rules\n\nUse this Skill before calling any Gecho TikTok MCP tool when the user asks to search TikTok, inspect a known video, find trending videos, analyze competitors, collect TikTok metadata, discover winning products, or research keyword trends.\n\nCore rules:\n\n- Use only the official Gecho MCP tools: `tiktok_search`, `tiktok_video`, `tiktok_influencer`, `tiktok_insight`, and `check_insight_status`.\n- Do not replace Gecho with WebSearch, browser automation, terminal scrapers, mcporter, unofficial APIs, or hand-written TikTok scraping.\n- Do not start more than one Gecho scraping or insight job in the same conversational turn.\n- Do not run Gecho scraping jobs in parallel because the workflow depends on one live browser tab and extension session.\n- If a tool fails, times out, or returns an error, stop and report the exact failure reason.\n- If `tiktok_search` returns no items, do not rewrite, translate, broaden, or retry the keyword automatically.\n- If the user requests video details but does not provide a video URL, ask for the exact URL instead of guessing or silently starting a keyword search.\n- If a known video is private, deleted, unavailable, or blocked, report that state instead of inventing details or comments.\n- If `tiktok_insight` starts successfully, report the `jobId` and explain that the user should check status later.\n- If `check_insight_status` says the job is still running, tell the user to wait before checking again.\n- If the official Gecho MCP tools are unavailable in the current session, provide setup instructions instead of probing the environment.\n- On first-run setup guidance, missing MCP tools, extension/session problems, timeouts, save failures, or any tool error, include the setup and support links block below.\n- Do not add the setup and support links block to normal successful search or insight responses unless the user asks for setup help.\n- Do not configure, edit, repair, or rewrite OpenClaw/Hermes/MCP settings on the user's behalf. Provide the setup commands and ask the user to run or approve them outside the tool workflow.\n- Do not use this Skill for TikTok Shop, X/Twitter, Amazon, or other platform workflows; those should have their own Gecho aggregate Skills.\n\nAllowed status behavior:\n\n- A user may explicitly ask to check one existing `jobId`; in that case, call `check_insight_status` once.\n- Do not start a new `tiktok_search` or `tiktok_insight` job in the same turn after checking status unless the user asks again in a later turn.\n\n## Standard workflows\n\n### Search workflow\n\n1. Use the exact keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_search`.\n4. If the result is empty, say that the exact keyword returned no results and stop.\n5. If results are present, summarize only the top 3 to 5 items and provide the saved file path.\n\n### Video detail workflow\n\n1. Confirm that the user supplied a TikTok video detail URL, normally containing `/video/`.\n2. Preserve the URL as provided and do not substitute another video.\n3. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable directory is available, omit `save_dir` and let Gecho use its default data directory.\n4. Pass the URL, a bounded `targetCount` when requested, and `save_dir` when reliable to `tiktok_video`.\n5. Wait for the single job to finish; do not run another Gecho scraping job in parallel.\n6. Summarize available video details and a representative sample of comments and replies, including the saved path when available.\n\n### Insight workflow\n\n1. Use the exact product, niche, or keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_insight`.\n4. Report the returned `jobId`.\n5. Tell the user the job may take several minutes and should be checked later with `check_insight_status`.\n\n### Status workflow\n\n1. Call `check_insight_status` with the provided `jobId`.\n2. If still running, report the current status and recommend waiting about 60 seconds.\n3. If completed, summarize the key findings and provide the saved file path.\n4. If failed, report the exact error and stop.\n\n## Setup and support links block\n\nUse this compact block on first-run guidance and all setup or failure responses. Keep it near the end of the answer, after the exact fix.\n\n````markdown\nHelpful Gecho links:\n\n- Website: https://gecho.ai/\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Support: Discord https://discord.gg/RFDVZMR6Tn, WeCom group QR https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg, 1:1 support QR https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n````\n\n## Setup-missing response\n\nUse this when the official Gecho MCP tools are not available, or when the user only installed the Skill and cannot run searches yet.\n\nUse the template below as the canonical setup-missing response. The agent may translate the prose to match the user's language, but must not summarize it into a shorter checklist unless the user explicitly asks for a short version.\n\nRequired content that must not be omitted:\n\n- State that Gecho Bridge MCP is not configured and the Skill alone cannot run TikTok.\n- Include all 3 requirements: MCP configured, Gecho Chrome extension installed and logged in, TikTok web logged in and tab kept open.\n- Include the OpenClaw MCP setup command and `openclaw mcp list` verification command.\n- Include the Chrome extension link.\n- Include after-setup wording that tells the user to return to OpenClaw Dashboard or Hermes and retry the request.\n- Include all related links: website, OpenClaw setup video, Hermes setup video, YouTube channel, GitHub/README, Discord, WeCom group QR code, and 1:1 support QR code.\n- Include related Gecho Skills listed in the template.\n- Do not recommend `openclaw plugins install` in this setup-missing response.\n\n````markdown\nGecho Bridge is not ready yet.\n\nThis Skill is installed, but the official Gecho Bridge MCP tools are not available in this session yet. Installing the Skill alone does not start the TikTok search service.\n\nGecho requires all 3 items below before TikTok search, video-detail collection, or insight can run:\n\n1. Gecho Bridge MCP is configured.\n2. The Gecho Chrome extension is installed and logged in to a Gecho account.\n3. TikTok web is logged in inside Chrome, with the TikTok tab kept open.\n\nFollow these first-time setup steps:\n\n**1. Install the Gecho Chrome extension**\n\nChrome Web Store:\nhttps://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb\n\nAfter installing, open the extension and log in to your Gecho account.\n\n**2. Configure Gecho Bridge MCP**\n\nOpenClaw MCP setup:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\nFor Hermes:\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n**3. Log in to TikTok web**\n\nOpen TikTok in Chrome, log in, and keep the logged-in TikTok tab open.\n\nAfter setup, return to OpenClaw Dashboard or Hermes and ask again, for example:\n\"Search computers on TikTok\" or provide a full TikTok video URL for detail collection.\n\n\n**Related links**\n\n- Website: https://gecho.ai/\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Discord support: https://discord.gg/RFDVZMR6Tn\n- WeCom group QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg\n- 1:1 support QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n\n**Related Gecho Skills**\n\n- `tiktok-video-search`: TikTok video search and metadata collection.\n- `tiktok-video`: detail data, comments, and replies for one known TikTok video URL.\n- `tiktok-insight`: TikTok product, trend, competitor, and content insight jobs.\n- `tiktok-search`: complete TikTok search and insight workflow.\n````\n\n## Troubleshooting\n\n| Situation | What to do |\n|---|---|\n| MCP tools are missing | Give the setup-missing response. Do not run local shell probes. |\n| User installed only the Skill | Explain that Skill-only install is not enough and provide the OpenClaw MCP setup command. |\n| Hermes MCP tools are missing | Provide the `hermes mcp add ...` command. Do not inspect or rewrite Hermes config files. |\n| Extension not connected | Ask the user to enable/login to the Gecho Chrome extension and keep a logged-in TikTok tab open. |\n| CAPTCHA or login wall | Ask the user to resolve it manually in Chrome, then retry in a later turn. |\n| Request timeout | Report the timeout, mention stuck page/network/CAPTCHA as likely causes, and stop. |\n| Empty search results | Say the exact keyword returned no results and ask the user to choose another keyword manually. |\n| Video URL is missing | Ask for the exact TikTok video detail URL; do not guess or switch to keyword search without the user's request. |\n| Video is private, deleted, or unavailable | Report the page state and ask the user to provide an accessible video URL if appropriate. |\n| Insight still running | Report running status and recommend checking again after about 60 seconds. |\n| Failed to save results | Ask the user to provide a valid absolute directory path with write permission. |\n\n## FAQ\n\n### Why is the Chrome extension required? Can't I just use the web page?\n\nGecho needs real-time platform data from a live browser session, such as TikTok videos and other platform data in Gecho workflows. The Chrome extension connects the AI workflow to the user's logged-in Chrome session; the Skill page alone cannot collect this data.\n\n### Why do I need to log in to TikTok? Can I use it without login?\n\nTikTok limits content access for logged-out users. After you log in, the extension can access the complete data available in your browser session, such as video captions/scripts, comments, engagement data, and other signals when available.\n\nGecho does not ask for or collect your TikTok password, private account information, or publish anything on your behalf.\n\n### Is this the same as TikTok video search?\n\nNo. `tiktok_search` discovers videos by keyword, while `tiktok_video` handles one known video URL and collects its detail data, comments, and replies. Use the matching single-tool Skill when the user wants only one of these focused workflows.\n\n### Need help?\n\nJoin the [WeCom group](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg) for community support, or scan the [1:1 support QR code](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg) for personal help.\n\n## Output guidelines\n\nFor successful search:\n\n- Say the search completed.\n- Include total result count if available.\n- Include saved file path if available.\n- Show top 3 to 5 results only.\n- Do not paste the full raw JSON into chat.\n\nFor successful video-detail collection:\n\n- Say that the requested TikTok video was opened and processed.\n- Summarize available identity and engagement fields without inventing missing values.\n- Report the number of comments and replies collected when available.\n- Show a bounded representative sample and include the saved path when available.\n\nFor successful insight start:\n\n- Say the insight job started.\n- Include the `jobId`.\n- Include expected saved path if available.\n- Tell the user to check status later.\n\nFor completed insight:\n\n- Summarize key findings.\n- Include saved file path.\n- Avoid claiming conclusions not supported by returned tool data.\n\nFor failures:\n\n- Report the exact tool error or failure state.\n- Provide only the relevant fix from Troubleshooting.\n- Include the setup and support links block so the user can continue setup through docs, videos, or support.\n- Do not retry in the same turn.\n\n## Scope and limits\n\nThis Skill should:\n\n- Help users complete the official Gecho setup when prerequisites are missing.\n- Route TikTok search and insight requests to the official Gecho MCP tools.\n- Route known TikTok video-detail requests to the official `tiktok_video` MCP tool.\n- Keep search, insight, and status-check flows explicit.\n- Summarize results without flooding the chat.\n\nThis Skill must never:\n\n- Pretend the Skill page alone is enough when MCP is missing.\n- Pretend `tiktok_insight` is synchronous.\n- Turn a missing video URL into an automatic keyword search.\n- Use unofficial TikTok scraping workflows.\n- Invent results when the tool returns no data.\n- Solve CAPTCHA, log in to TikTok, or operate the user's browser outside the official Gecho MCP workflow.\n\nFile v1.1.38:_meta.json\n\n{\n  \"ownerId\": \"kn7b732kkfejrjfkjq2h0hc9qn84nv5k\",\n  \"slug\": \"tiktok-search\",\n  \"version\": \"1.1.38\",\n  \"publishedAt\": 1790153928634\n}\n\nFile v1.1.38:skill-card.md\n\n## Description:\n\nSearch TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gecho-ai](https://clawhub.ai/user/gecho-ai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users, developers, and social-media researchers use this skill to search TikTok, collect video and creator metadata, inspect one known video, and run product, trend, competitor, or content insight workflows through Gecho Bridge MCP.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The workflow depends on a third-party MCP server and Chrome extension using the user's logged-in TikTok browser session.\n\nMitigation: Install and use it only when comfortable allowing that integration to read TikTok pages available in the browser session.\n\nRisk: Search, video, and insight workflows can fail when Gecho Bridge MCP, the Chrome extension, or the TikTok browser session is not ready.\n\nMitigation: Confirm Node/npx, Gecho Bridge MCP, the logged-in Gecho Chrome extension, and an active logged-in TikTok tab before use.\n\nRisk: Collected TikTok result sets may be saved locally.\n\nMitigation: Choose an appropriate save directory and handle generated files according to the user's data-handling requirements.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gecho-ai/skills/tiktok-search)\n- [Gecho website](https://gecho.ai/)\n- [Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n- [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- [OpenClaw setup video](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- [Hermes setup video](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown with concise summaries, setup commands, saved file paths, and bounded result samples]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May reference JSON result files produced by Gecho MCP tools; successful insight starts return a jobId for later status checks.]\n\n## Skill Version(s):\n\n1.1.38 (source: server release metadata and artifact _meta.json)\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 v1.1.37: 3 files, 8877 bytes\n\nFiles: _meta.json (133b), skill-card.md (2907b), SKILL.md (22727b)\n\nFile v1.1.37:SKILL.md\n\n---\nname: tiktok-search\ndescription: Search TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server.\nmetadata:\n  openclaw:\n    os: [\"darwin\", \"linux\", \"win32\"]\n    requires:\n      bins: [\"node\", \"npx\"]\n  hermes:\n    tags: [tiktok, search, insight, gecho, mcp]\n    category: social-media\n    os: [darwin, linux, windows]\n---\n\n# TikTok Search & Insight by Gecho\n\nSearch TikTok from an AI chat, collect structured video metadata, and run async product or trend insight jobs through the official Gecho Bridge MCP workflow.\n\nThis is the default TikTok aggregate Skill for Gecho. It covers TikTok video search, creator collection, one-video detail and comment collection, and TikTok insight. Single-tool TikTok Skills may exist for distribution and search traffic, but this Skill is the recommended default for users who want the complete TikTok research workflow.\n\n## Critical prerequisite: read before use\n\nGecho Skills must be used with the Gecho Chrome extension. You must be logged in to both your Gecho account in the extension and TikTok in the Chrome web app. If either login is missing, all TikTok search and insight features may fail even though the Skill is installed.\n\n## 3-step quick start\n\n### Step 1: Install the Gecho Chrome extension\n\n1. Open the [Gecho Chrome extension download page](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb).\n2. Click `Add to Chrome`, then confirm `Add extension`.\n\n### Step 2: Log in to the Gecho extension\n\nOpen the Gecho extension in Chrome and log in to your Gecho account. Keep the extension online.\n\n### Step 3: Log in to TikTok web\n\nOpen TikTok in Chrome and log in to the TikTok web app. Keep the logged-in TikTok tab open.\n\nAfter setup is complete, return to the OpenClaw dashboard or Hermes and ask: \"Search xxx on TikTok\". For example: \"Search hamburgers on TikTok\".\n\n## Official links and setup help\n\n- Website: [gecho.ai](https://gecho.ai/)\n- GitHub: [gecho-ai/gecho-bridge](https://github.com/gecho-ai/gecho-bridge)\n- YouTube channel: [@Gecho-AI](https://www.youtube.com/@Gecho-AI)\n- Chrome extension: [Gecho Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- OpenClaw setup video: [OpenClaw + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- Hermes setup video: [Hermes + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n- Discord: [https://discord.gg/RFDVZMR6Tn](https://discord.gg/RFDVZMR6Tn)\n- WeCom group QR code: [qywx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg)\n- 1:1 support QR code: [wx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg)\n\n## What this skill does\n\n- Finds high-performing TikTok videos for a keyword.\n- Collects titles, authors, engagement data, and video links.\n- Retrieves detail data, comments, and replies for one known TikTok video URL.\n- Saves full raw result sets to a local JSON file.\n- Starts async insight jobs for product research, competitor analysis, and trend discovery.\n\nBest-fit prompts:\n\n- \"Search TikTok for portable blender and show the top liked videos.\"\n- \"Find winning hooks for cat toy videos.\"\n- \"Run product opportunity insight for outdoor picnic mat.\"\n- \"Check the status of my previous TikTok insight job.\"\n- \"Get the details and comments for this TikTok video: https://www.tiktok.com/@user/video/123.\"\n\n## Important: Skill-only install is not enough\n\nThis Skill is the instruction layer. It tells the AI when and how to use Gecho.\n\nTo actually run TikTok searches, video-detail collection, or insight jobs, the user also needs:\n\n- the Gecho Bridge MCP server\n- the Gecho Chrome extension\n- Chrome with TikTok logged in\n- the Gecho extension logged in to a Gecho account and online\n\nIf the user installed only this Skill from ClawHub, searches will not work until the Gecho Bridge MCP server is configured. In that case, use the MCP setup path below.\n\nAlready installed Gecho Bridge? If `@gecho-ai/gecho-bridge-bundle` is installed and the Gecho MCP tools are visible, no extra MCP setup is needed for this Skill.\n\n## Quick start\n\n### OpenClaw Skill install: configure MCP\n\nIf this Skill is already installed in OpenClaw, configure the Gecho Bridge MCP server once:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\n### Optional: OpenClaw Bundle Plugin\n\nIf the user has not installed this Skill yet and prefers plugin management, the bundle plugin can install Gecho with the MCP entry configured:\n\n```bash\nopenclaw plugins install clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\nTo upgrade later:\n\n```bash\nopenclaw plugins update clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\n### Hermes setup\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\nIf Hermes cannot find `npx` even though Node.js is installed, use the absolute `npx` path. On many macOS Homebrew installs this is:\n\n```bash\nhermes mcp add gecho-bridge --command /opt/homebrew/bin/npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n## First-run checklist\n\nBefore the first search, make sure:\n\n- Node.js `>= 18` is available.\n- The [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) is installed.\n- Chrome is open with TikTok logged in.\n- The Gecho extension is logged in to a Gecho account and online.\n- The TikTok tab is not blocked by CAPTCHA, login walls, or a frozen page.\n\nFull setup guide:\n[Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n\nFor videos and support links, see the official links section above.\n\n## Tool choice\n\n| User goal | Use tool | Notes |\n|---|---|---|\n| Search TikTok videos and collect metadata | `tiktok_search` | Returns results directly |\n| Collect detail data, comments, and replies from one known video | `tiktok_video` | Requires a TikTok video URL |\n| Collect videos from a TikTok creator | `tiktok_influencer` | Uses the creator's `uniqueId` |\n| Analyze a niche, product, trend, or competitor opportunity | `tiktok_insight` | Starts an async job and returns a `jobId` |\n| Check an existing async insight job | `check_insight_status` | Use the `jobId` returned by `tiktok_insight` |\n\nUse this aggregate Skill for broad TikTok research requests. If the user explicitly asks for one exact raw tool only, such as \"run tiktok_video\", a matching single-tool distribution Skill may be used instead when available.\n\n## Official MCP tools\n\n### `tiktok_search`\n\nSearches TikTok for a keyword, scrolls the page through the Gecho browser extension, returns structured metadata, and saves the full result set.\n\nParameters:\n\n- `query` string, required: search keyword or phrase.\n- `targetCount` number, optional: requested result count; defaults to `100`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A JSON array of video metadata, plus a saved local file path when results are written successfully.\n\n### `tiktok_video`\n\nGets TikTok video detail data and comments. It opens the video detail page, expands comments, and continues scrolling to collect comments and replies.\n\nParameters:\n\n- `url` string, required: TikTok video detail-page URL, for example `https://www.tiktok.com/@user/video/123...`.\n- `targetCount` number, optional: maximum number of comments and replies to collect; default and maximum are `200`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- Structured video detail data, comments, and replies, plus a saved local file path when results are written successfully.\n\n### `tiktok_influencer`\n\nCollects videos published by one TikTok creator.\n\nParameters:\n\n- `uniqueId` string, required: the creator's TikTok `uniqueId` (for example, `zachking`).\n- `targetCount` number, optional: requested video count; defaults to `100`. Keep requests at or below `500`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\n### `tiktok_insight`\n\nStarts an asynchronous TikTok insight job for product, trend, or competitor research.\n\nParameters:\n\n- `query` string, required: search keyword or product/category phrase.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A `jobId`. The final result must be checked later with `check_insight_status`.\n\n### `check_insight_status`\n\nChecks the status or final result of an existing insight job.\n\nParameters:\n\n- `jobId` string, required: the job ID returned by `tiktok_insight`.\n\nExpected result:\n\n- `running`, `error`, or completed insight data.\n\n## Agent execution rules\n\nUse this Skill before calling any Gecho TikTok MCP tool when the user asks to search TikTok, inspect a known video, find trending videos, analyze competitors, collect TikTok metadata, discover winning products, or research keyword trends.\n\nCore rules:\n\n- Use only the official Gecho MCP tools: `tiktok_search`, `tiktok_video`, `tiktok_influencer`, `tiktok_insight`, and `check_insight_status`.\n- Do not replace Gecho with WebSearch, browser automation, terminal scrapers, mcporter, unofficial APIs, or hand-written TikTok scraping.\n- Do not start more than one Gecho scraping or insight job in the same conversational turn.\n- Do not run Gecho scraping jobs in parallel because the workflow depends on one live browser tab and extension session.\n- If a tool fails, times out, or returns an error, stop and report the exact failure reason.\n- If `tiktok_search` returns no items, do not rewrite, translate, broaden, or retry the keyword automatically.\n- If the user requests video details but does not provide a video URL, ask for the exact URL instead of guessing or silently starting a keyword search.\n- If a known video is private, deleted, unavailable, or blocked, report that state instead of inventing details or comments.\n- If `tiktok_insight` starts successfully, report the `jobId` and explain that the user should check status later.\n- If `check_insight_status` says the job is still running, tell the user to wait before checking again.\n- If the official Gecho MCP tools are unavailable in the current session, provide setup instructions instead of probing the environment.\n- On first-run setup guidance, missing MCP tools, extension/session problems, timeouts, save failures, or any tool error, include the setup and support links block below.\n- Do not add the setup and support links block to normal successful search or insight responses unless the user asks for setup help.\n- Do not configure, edit, repair, or rewrite OpenClaw/Hermes/MCP settings on the user's behalf. Provide the setup commands and ask the user to run or approve them outside the tool workflow.\n- Do not use this Skill for TikTok Shop, X/Twitter, Amazon, or other platform workflows; those should have their own Gecho aggregate Skills.\n\nAllowed status behavior:\n\n- A user may explicitly ask to check one existing `jobId`; in that case, call `check_insight_status` once.\n- Do not start a new `tiktok_search` or `tiktok_insight` job in the same turn after checking status unless the user asks again in a later turn.\n\n## Standard workflows\n\n### Search workflow\n\n1. Use the exact keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_search`.\n4. If the result is empty, say that the exact keyword returned no results and stop.\n5. If results are present, summarize only the top 3 to 5 items and provide the saved file path.\n\n### Video detail workflow\n\n1. Confirm that the user supplied a TikTok video detail URL, normally containing `/video/`.\n2. Preserve the URL as provided and do not substitute another video.\n3. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable directory is available, omit `save_dir` and let Gecho use its default data directory.\n4. Pass the URL, a bounded `targetCount` when requested, and `save_dir` when reliable to `tiktok_video`.\n5. Wait for the single job to finish; do not run another Gecho scraping job in parallel.\n6. Summarize available video details and a representative sample of comments and replies, including the saved path when available.\n\n### Insight workflow\n\n1. Use the exact product, niche, or keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_insight`.\n4. Report the returned `jobId`.\n5. Tell the user the job may take several minutes and should be checked later with `check_insight_status`.\n\n### Status workflow\n\n1. Call `check_insight_status` with the provided `jobId`.\n2. If still running, report the current status and recommend waiting about 60 seconds.\n3. If completed, summarize the key findings and provide the saved file path.\n4. If failed, report the exact error and stop.\n\n## Setup and support links block\n\nUse this compact block on first-run guidance and all setup or failure responses. Keep it near the end of the answer, after the exact fix.\n\n````markdown\nHelpful Gecho links:\n\n- Website: https://gecho.ai/\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Support: Discord https://discord.gg/RFDVZMR6Tn, WeCom group QR https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg, 1:1 support QR https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n````\n\n## Setup-missing response\n\nUse this when the official Gecho MCP tools are not available, or when the user only installed the Skill and cannot run searches yet.\n\nUse the template below as the canonical setup-missing response. The agent may translate the prose to match the user's language, but must not summarize it into a shorter checklist unless the user explicitly asks for a short version.\n\nRequired content that must not be omitted:\n\n- State that Gecho Bridge MCP is not configured and the Skill alone cannot run TikTok.\n- Include all 3 requirements: MCP configured, Gecho Chrome extension installed and logged in, TikTok web logged in and tab kept open.\n- Include the OpenClaw MCP setup command and `openclaw mcp list` verification command.\n- Include the Chrome extension link.\n- Include after-setup wording that tells the user to return to OpenClaw Dashboard or Hermes and retry the request.\n- Include all related links: website, OpenClaw setup video, Hermes setup video, YouTube channel, GitHub/README, Discord, WeCom group QR code, and 1:1 support QR code.\n- Include related Gecho Skills listed in the template.\n- Do not recommend `openclaw plugins install` in this setup-missing response.\n\n````markdown\nGecho Bridge is not ready yet.\n\nThis Skill is installed, but the official Gecho Bridge MCP tools are not available in this session yet. Installing the Skill alone does not start the TikTok search service.\n\nGecho requires all 3 items below before TikTok search, video-detail collection, or insight can run:\n\n1. Gecho Bridge MCP is configured.\n2. The Gecho Chrome extension is installed and logged in to a Gecho account.\n3. TikTok web is logged in inside Chrome, with the TikTok tab kept open.\n\nFollow these first-time setup steps:\n\n**1. Install the Gecho Chrome extension**\n\nChrome Web Store:\nhttps://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb\n\nAfter installing, open the extension and log in to your Gecho account.\n\n**2. Configure Gecho Bridge MCP**\n\nOpenClaw MCP setup:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\nFor Hermes:\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n**3. Log in to TikTok web**\n\nOpen TikTok in Chrome, log in, and keep the logged-in TikTok tab open.\n\nAfter setup, return to OpenClaw Dashboard or Hermes and ask again, for example:\n\"Search computers on TikTok\" or provide a full TikTok video URL for detail collection.\n\n\n**Related links**\n\n- Website: https://gecho.ai/\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Discord support: https://discord.gg/RFDVZMR6Tn\n- WeCom group QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg\n- 1:1 support QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n\n**Related Gecho Skills**\n\n- `tiktok-video-search`: TikTok video search and metadata collection.\n- `tiktok-video`: detail data, comments, and replies for one known TikTok video URL.\n- `tiktok-insight`: TikTok product, trend, competitor, and content insight jobs.\n- `tiktok-search`: complete TikTok search and insight workflow.\n````\n\n## Troubleshooting\n\n| Situation | What to do |\n|---|---|\n| MCP tools are missing | Give the setup-missing response. Do not run local shell probes. |\n| User installed only the Skill | Explain that Skill-only install is not enough and provide the OpenClaw MCP setup command. |\n| Hermes MCP tools are missing | Provide the `hermes mcp add ...` command. Do not inspect or rewrite Hermes config files. |\n| Extension not connected | Ask the user to enable/login to the Gecho Chrome extension and keep a logged-in TikTok tab open. |\n| CAPTCHA or login wall | Ask the user to resolve it manually in Chrome, then retry in a later turn. |\n| Request timeout | Report the timeout, mention stuck page/network/CAPTCHA as likely causes, and stop. |\n| Empty search results | Say the exact keyword returned no results and ask the user to choose another keyword manually. |\n| Video URL is missing | Ask for the exact TikTok video detail URL; do not guess or switch to keyword search without the user's request. |\n| Video is private, deleted, or unavailable | Report the page state and ask the user to provide an accessible video URL if appropriate. |\n| Insight still running | Report running status and recommend checking again after about 60 seconds. |\n| Failed to save results | Ask the user to provide a valid absolute directory path with write permission. |\n\n## FAQ\n\n### Why is the Chrome extension required? Can't I just use the web page?\n\nGecho needs real-time platform data from a live browser session, such as TikTok videos and other platform data in Gecho workflows. The Chrome extension connects the AI workflow to the user's logged-in Chrome session; the Skill page alone cannot collect this data.\n\n### Why do I need to log in to TikTok? Can I use it without login?\n\nTikTok limits content access for logged-out users. After you log in, the extension can access the complete data available in your browser session, such as video captions/scripts, comments, engagement data, and other signals when available.\n\nGecho does not ask for or collect your TikTok password, private account information, or publish anything on your behalf.\n\n### Is this the same as TikTok video search?\n\nNo. `tiktok_search` discovers videos by keyword, while `tiktok_video` handles one known video URL and collects its detail data, comments, and replies. Use the matching single-tool Skill when the user wants only one of these focused workflows.\n\n### Need help?\n\nJoin the [WeCom group](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg) for community support, or scan the [1:1 support QR code](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg) for personal help.\n\n## Output guidelines\n\nFor successful search:\n\n- Say the search completed.\n- Include total result count if available.\n- Include saved file path if available.\n- Show top 3 to 5 results only.\n- Do not paste the full raw JSON into chat.\n\nFor successful video-detail collection:\n\n- Say that the requested TikTok video was opened and processed.\n- Summarize available identity and engagement fields without inventing missing values.\n- Report the number of comments and replies collected when available.\n- Show a bounded representative sample and include the saved path when available.\n\nFor successful insight start:\n\n- Say the insight job started.\n- Include the `jobId`.\n- Include expected saved path if available.\n- Tell the user to check status later.\n\nFor completed insight:\n\n- Summarize key findings.\n- Include saved file path.\n- Avoid claiming conclusions not supported by returned tool data.\n\nFor failures:\n\n- Report the exact tool error or failure state.\n- Provide only the relevant fix from Troubleshooting.\n- Include the setup and support links block so the user can continue setup through docs, videos, or support.\n- Do not retry in the same turn.\n\n## Scope and limits\n\nThis Skill should:\n\n- Help users complete the official Gecho setup when prerequisites are missing.\n- Route TikTok search and insight requests to the official Gecho MCP tools.\n- Route known TikTok video-detail requests to the official `tiktok_video` MCP tool.\n- Keep search, insight, and status-check flows explicit.\n- Summarize results without flooding the chat.\n\nThis Skill must never:\n\n- Pretend the Skill page alone is enough when MCP is missing.\n- Pretend `tiktok_insight` is synchronous.\n- Turn a missing video URL into an automatic keyword search.\n- Use unofficial TikTok scraping workflows.\n- Invent results when the tool returns no data.\n- Solve CAPTCHA, log in to TikTok, or operate the user's browser outside the official Gecho MCP workflow.\n\nFile v1.1.37:_meta.json\n\n{\n  \"ownerId\": \"kn7b732kkfejrjfkjq2h0hc9qn84nv5k\",\n  \"slug\": \"tiktok-search\",\n  \"version\": \"1.1.37\",\n  \"publishedAt\": 1787324702837\n}\n\nFile v1.1.37:skill-card.md\n\n## Description:\n\nSearch TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gecho-ai](https://clawhub.ai/user/gecho-ai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users, social media researchers, and commerce teams use this skill to search TikTok, inspect known videos, collect creator metadata, and start TikTok product, trend, competitor, and content insight jobs through Gecho Bridge.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The setup uses an unpinned npm package, so the installed Gecho Bridge MCP code can change outside the reviewed artifact.\n\nMitigation: Review the Gecho Bridge MCP package before installation and prefer a pinned, audited package version instead of @latest.\n\nRisk: The workflow uses the user's logged-in Chrome and TikTok session and may save TikTok metadata, comments, and replies locally.\n\nMitigation: Run the workflow with the least privileges practical, use an appropriate TikTok account/session, and store exported result files only in controlled locations.\n\nRisk: Search, detail, and insight features depend on Gecho Bridge MCP, the Gecho Chrome extension, Chrome, and active TikTok and Gecho logins.\n\nMitigation: Verify the MCP server, extension login, TikTok web login, and browser state before use; resolve CAPTCHAs or login walls manually before retrying.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gecho-ai/skills/tiktok-search)\n- [Gecho website](https://gecho.ai/)\n- [Gecho Bridge repository](https://github.com/gecho-ai/gecho-bridge)\n- [Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n- [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- [OpenClaw setup video](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- [Hermes setup video](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, json, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown responses with structured TikTok result summaries, JSON data references, and inline shell commands when setup is needed]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May include saved local JSON result paths, TikTok metadata, comment summaries, setup commands, and asynchronous insight job IDs.]\n\n## Skill Version(s):\n\n1.1.37 (source: server release metadata and artifact _meta.json)\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 v1.1.36: 3 files, 7899 bytes\n\nFiles: _meta.json (133b), skill-card.md (2554b), SKILL.md (19129b)\n\nFile v1.1.36:SKILL.md\n\n---\nname: tiktok-search\ndescription: Search TikTok videos, collect creator videos, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server.\nmetadata:\n  openclaw:\n    os: [\"darwin\", \"linux\", \"win32\"]\n    requires:\n      bins: [\"node\", \"npx\"]\n  hermes:\n    tags: [tiktok, search, insight, gecho, mcp]\n    category: social-media\n    os: [darwin, linux, windows]\n---\n\n# TikTok Search & Insight by Gecho\n\nSearch TikTok from an AI chat, collect structured video metadata, and run async product or trend insight jobs through the official Gecho Bridge MCP workflow.\n\nThis is the default TikTok aggregate Skill for Gecho. It covers TikTok video search, creator collection, and TikTok insight. Single-tool TikTok Skills may exist for distribution and search traffic, but this Skill is the recommended default for users who want the complete TikTok research workflow.\n\n## Critical prerequisite: read before use\n\nGecho Skills must be used with the Gecho Chrome extension. You must be logged in to both your Gecho account in the extension and TikTok in the Chrome web app. If either login is missing, all TikTok search and insight features may fail even though the Skill is installed.\n\n## 3-step quick start\n\n### Step 1: Install the Gecho Chrome extension\n\n1. Open the [Gecho Chrome extension download page](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb).\n2. Click `Add to Chrome`, then confirm `Add extension`.\n\n### Step 2: Log in to the Gecho extension\n\nOpen the Gecho extension in Chrome and log in to your Gecho account. Keep the extension online.\n\n### Step 3: Log in to TikTok web\n\nOpen TikTok in Chrome and log in to the TikTok web app. Keep the logged-in TikTok tab open.\n\nAfter setup is complete, return to the OpenClaw dashboard or Hermes and ask: \"Search xxx on TikTok\". For example: \"Search hamburgers on TikTok\".\n\n## Official links and setup help\n\n- Website: [gecho.ai](https://gecho.ai/)\n- GitHub: [gecho-ai/gecho-bridge](https://github.com/gecho-ai/gecho-bridge)\n- YouTube channel: [@Gecho-AI](https://www.youtube.com/@Gecho-AI)\n- Chrome extension: [Gecho Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- OpenClaw setup video: [OpenClaw + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- Hermes setup video: [Hermes + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n- Discord: [https://discord.gg/RFDVZMR6Tn](https://discord.gg/RFDVZMR6Tn)\n- WeCom group QR code: [qywx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg)\n- 1:1 support QR code: [wx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg)\n\n## What this skill does\n\n- Finds high-performing TikTok videos for a keyword.\n- Collects titles, authors, engagement data, and video links.\n- Saves full raw result sets to a local JSON file.\n- Starts async insight jobs for product research, competitor analysis, and trend discovery.\n\nBest-fit prompts:\n\n- \"Search TikTok for portable blender and show the top liked videos.\"\n- \"Find winning hooks for cat toy videos.\"\n- \"Run product opportunity insight for outdoor picnic mat.\"\n- \"Check the status of my previous TikTok insight job.\"\n\n## Important: Skill-only install is not enough\n\nThis Skill is the instruction layer. It tells the AI when and how to use Gecho.\n\nTo actually run TikTok searches, the user also needs:\n\n- the Gecho Bridge MCP server\n- the Gecho Chrome extension\n- Chrome with TikTok logged in\n- the Gecho extension logged in to a Gecho account and online\n\nIf the user installed only this Skill from ClawHub, searches will not work until the Gecho Bridge MCP server is configured. In that case, use the MCP setup path below.\n\nAlready installed Gecho Bridge? If `@gecho-ai/gecho-bridge-bundle` is installed and the Gecho MCP tools are visible, no extra MCP setup is needed for this Skill.\n\n## Quick start\n\n### OpenClaw Skill install: configure MCP\n\nIf this Skill is already installed in OpenClaw, configure the Gecho Bridge MCP server once:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\n### Optional: OpenClaw Bundle Plugin\n\nIf the user has not installed this Skill yet and prefers plugin management, the bundle plugin can install Gecho with the MCP entry configured:\n\n```bash\nopenclaw plugins install clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\nTo upgrade later:\n\n```bash\nopenclaw plugins update clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\n### Hermes setup\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\nIf Hermes cannot find `npx` even though Node.js is installed, use the absolute `npx` path. On many macOS Homebrew installs this is:\n\n```bash\nhermes mcp add gecho-bridge --command /opt/homebrew/bin/npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n## First-run checklist\n\nBefore the first search, make sure:\n\n- Node.js `>= 18` is available.\n- The [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) is installed.\n- Chrome is open with TikTok logged in.\n- The Gecho extension is logged in to a Gecho account and online.\n- The TikTok tab is not blocked by CAPTCHA, login walls, or a frozen page.\n\nFull setup guide:\n[Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n\nFor videos and support links, see the official links section above.\n\n## Tool choice\n\n| User goal | Use tool | Notes |\n|---|---|---|\n| Search TikTok videos and collect metadata | `tiktok_search` | Returns results directly |\n| Collect videos from a TikTok creator | `tiktok_influencer` | Uses the creator's `uniqueId` |\n| Analyze a niche, product, trend, or competitor opportunity | `tiktok_insight` | Starts an async job and returns a `jobId` |\n| Check an existing async insight job | `check_insight_status` | Use the `jobId` returned by `tiktok_insight` |\n\nUse this aggregate Skill for broad TikTok research requests. If the user explicitly asks for one exact raw tool only, such as \"run tiktok_search\", a matching single-tool distribution Skill may be used instead when available.\n\n## Official MCP tools\n\n### `tiktok_search`\n\nSearches TikTok for a keyword, scrolls the page through the Gecho browser extension, returns structured metadata, and saves the full result set.\n\nParameters:\n\n- `query` string, required: search keyword or phrase.\n- `targetCount` number, optional: requested result count; defaults to `100`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A JSON array of video metadata, plus a saved local file path when results are written successfully.\n\n### `tiktok_influencer`\n\nCollects videos published by one TikTok creator.\n\nParameters:\n\n- `uniqueId` string, required: the creator's TikTok `uniqueId` (for example, `zachking`).\n- `targetCount` number, optional: requested video count; defaults to `100`. Keep requests at or below `500`.\n\n### `tiktok_insight`\n\nStarts an asynchronous TikTok insight job for product, trend, or competitor research.\n\nParameters:\n\n- `query` string, required: search keyword or product/category phrase.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A `jobId`. The final result must be checked later with `check_insight_status`.\n\n### `check_insight_status`\n\nChecks the status or final result of an existing insight job.\n\nParameters:\n\n- `jobId` string, required: the job ID returned by `tiktok_insight`.\n\nExpected result:\n\n- `running`, `error`, or completed insight data.\n\n## Agent execution rules\n\nUse this Skill before calling any Gecho TikTok MCP tool when the user asks to search TikTok, find trending videos, analyze competitors, collect TikTok metadata, discover winning products, or research keyword trends.\n\nCore rules:\n\n- Use only the official Gecho MCP tools: `tiktok_search`, `tiktok_influencer`, `tiktok_insight`, and `check_insight_status`.\n- Do not replace Gecho with WebSearch, browser automation, terminal scrapers, mcporter, unofficial APIs, or hand-written TikTok scraping.\n- Do not start more than one Gecho scraping or insight job in the same conversational turn.\n- Do not run Gecho scraping jobs in parallel because the workflow depends on one live browser tab and extension session.\n- If a tool fails, times out, or returns an error, stop and report the exact failure reason.\n- If `tiktok_search` returns no items, do not rewrite, translate, broaden, or retry the keyword automatically.\n- If `tiktok_insight` starts successfully, report the `jobId` and explain that the user should check status later.\n- If `check_insight_status` says the job is still running, tell the user to wait before checking again.\n- If the official Gecho MCP tools are unavailable in the current session, provide setup instructions instead of probing the environment.\n- On first-run setup guidance, missing MCP tools, extension/session problems, timeouts, save failures, or any tool error, include the setup and support links block below.\n- Do not add the setup and support links block to normal successful search or insight responses unless the user asks for setup help.\n- Do not configure, edit, repair, or rewrite OpenClaw/Hermes/MCP settings on the user's behalf. Provide the setup commands and ask the user to run or approve them outside the tool workflow.\n- Do not use this Skill for TikTok Shop, X/Twitter, Amazon, or other platform workflows; those should have their own Gecho aggregate Skills.\n\nAllowed status behavior:\n\n- A user may explicitly ask to check one existing `jobId`; in that case, call `check_insight_status` once.\n- Do not start a new `tiktok_search` or `tiktok_insight` job in the same turn after checking status unless the user asks again in a later turn.\n\n## Standard workflows\n\n### Search workflow\n\n1. Use the exact keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_search`.\n4. If the result is empty, say that the exact keyword returned no results and stop.\n5. If results are present, summarize only the top 3 to 5 items and provide the saved file path.\n\n### Insight workflow\n\n1. Use the exact product, niche, or keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_insight`.\n4. Report the returned `jobId`.\n5. Tell the user the job may take several minutes and should be checked later with `check_insight_status`.\n\n### Status workflow\n\n1. Call `check_insight_status` with the provided `jobId`.\n2. If still running, report the current status and recommend waiting about 60 seconds.\n3. If completed, summarize the key findings and provide the saved file path.\n4. If failed, report the exact error and stop.\n\n## Setup and support links block\n\nUse this compact block on first-run guidance and all setup or failure responses. Keep it near the end of the answer, after the exact fix.\n\n````markdown\nHelpful Gecho links:\n\n- Website: https://gecho.ai/\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Support: Discord https://discord.gg/RFDVZMR6Tn, WeCom group QR https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg, 1:1 support QR https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n````\n\n## Setup-missing response\n\nUse this when the official Gecho MCP tools are not available, or when the user only installed the Skill and cannot run searches yet.\n\nUse the template below as the canonical setup-missing response. The agent may translate the prose to match the user's language, but must not summarize it into a shorter checklist unless the user explicitly asks for a short version.\n\nRequired content that must not be omitted:\n\n- State that Gecho Bridge MCP is not configured and the Skill alone cannot run TikTok.\n- Include all 3 requirements: MCP configured, Gecho Chrome extension installed and logged in, TikTok web logged in and tab kept open.\n- Include the OpenClaw MCP setup command and `openclaw mcp list` verification command.\n- Include the Chrome extension link.\n- Include after-setup wording that tells the user to return to OpenClaw Dashboard or Hermes and retry the request.\n- Include all related links: website, OpenClaw setup video, Hermes setup video, YouTube channel, GitHub/README, Discord, WeCom group QR code, and 1:1 support QR code.\n- Include related Gecho Skills listed in the template.\n- Do not recommend `openclaw plugins install` in this setup-missing response.\n\n````markdown\nGecho Bridge is not ready yet.\n\nThis Skill is installed, but the official Gecho Bridge MCP tools are not available in this session yet. Installing the Skill alone does not start the TikTok search service.\n\nGecho requires all 3 items below before TikTok search or insight can run:\n\n1. Gecho Bridge MCP is configured.\n2. The Gecho Chrome extension is installed and logged in to a Gecho account.\n3. TikTok web is logged in inside Chrome, with the TikTok tab kept open.\n\nFollow these first-time setup steps:\n\n**1. Install the Gecho Chrome extension**\n\nChrome Web Store:\nhttps://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb\n\nAfter installing, open the extension and log in to your Gecho account.\n\n**2. Configure Gecho Bridge MCP**\n\nOpenClaw MCP setup:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\nFor Hermes:\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n**3. Log in to TikTok web**\n\nOpen TikTok in Chrome, log in, and keep the logged-in TikTok tab open.\n\nAfter setup, return to OpenClaw Dashboard or Hermes and ask again, for example:\n\"Search computers on TikTok\"\n\n\n**Related links**\n\n- Website: https://gecho.ai/\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Discord support: https://discord.gg/RFDVZMR6Tn\n- WeCom group QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg\n- 1:1 support QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n\n**Related Gecho Skills**\n\n- `tiktok-video-search`: TikTok video search and metadata collection.\n- `tiktok-insight`: TikTok product, trend, competitor, and content insight jobs.\n- `tiktok-search`: complete TikTok search and insight workflow.\n````\n\n## Troubleshooting\n\n| Situation | What to do |\n|---|---|\n| MCP tools are missing | Give the setup-missing response. Do not run local shell probes. |\n| User installed only the Skill | Explain that Skill-only install is not enough and provide the OpenClaw MCP setup command. |\n| Hermes MCP tools are missing | Provide the `hermes mcp add ...` command. Do not inspect or rewrite Hermes config files. |\n| Extension not connected | Ask the user to enable/login to the Gecho Chrome extension and keep a logged-in TikTok tab open. |\n| CAPTCHA or login wall | Ask the user to resolve it manually in Chrome, then retry in a later turn. |\n| Request timeout | Report the timeout, mention stuck page/network/CAPTCHA as likely causes, and stop. |\n| Empty search results | Say the exact keyword returned no results and ask the user to choose another keyword manually. |\n| Insight still running | Report running status and recommend checking again after about 60 seconds. |\n| Failed to save results | Ask the user to provide a valid absolute directory path with write permission. |\n\n## FAQ\n\n### Why is the Chrome extension required? Can't I just use the web page?\n\nGecho needs real-time platform data from a live browser session, such as TikTok videos and other platform data in Gecho workflows. The Chrome extension connects the AI workflow to the user's logged-in Chrome session; the Skill page alone cannot collect this data.\n\n### Why do I need to log in to TikTok? Can I use it without login?\n\nTikTok limits content access for logged-out users. After you log in, the extension can access the complete data available in your browser session, such as video captions/scripts, comments, engagement data, and other signals when available.\n\nGecho does not ask for or collect your TikTok password, private account information, or publish anything on your behalf.\n\n### Need help?\n\nJoin the [WeCom group](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg) for community support, or scan the [1:1 support QR code](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg) for personal help.\n\n## Output guidelines\n\nFor successful search:\n\n- Say the search completed.\n- Include total result count if available.\n- Include saved file path if available.\n- Show top 3 to 5 results only.\n- Do not paste the full raw JSON into chat.\n\nFor successful insight start:\n\n- Say the insight job started.\n- Include the `jobId`.\n- Include expected saved path if available.\n- Tell the user to check status later.\n\nFor completed insight:\n\n- Summarize key findings.\n- Include saved file path.\n- Avoid claiming conclusions not supported by returned tool data.\n\nFor failures:\n\n- Report the exact tool error or failure state.\n- Provide only the relevant fix from Troubleshooting.\n- Include the setup and support links block so the user can continue setup through docs, videos, or support.\n- Do not retry in the same turn.\n\n## Scope and limits\n\nThis Skill should:\n\n- Help users complete the official Gecho setup when prerequisites are missing.\n- Route TikTok search and insight requests to the official Gecho MCP tools.\n- Keep search, insight, and status-check flows explicit.\n- Summarize results without flooding the chat.\n\nThis Skill must never:\n\n- Pretend the Skill page alone is enough when MCP is missing.\n- Pretend `tiktok_insight` is synchronous.\n- Use unofficial TikTok scraping workflows.\n- Invent results when the tool returns no data.\n- Solve CAPTCHA, log in to TikTok, or operate the user's browser outside the official Gecho MCP workflow.\n\nFile v1.1.36:_meta.json\n\n{\n  \"ownerId\": \"kn7b732kkfejrjfkjq2h0hc9qn84nv5k\",\n  \"slug\": \"tiktok-search\",\n  \"version\": \"1.1.36\",\n  \"publishedAt\": 1786856945902\n}\n\nFile v1.1.36:skill-card.md\n\n## Description:\n\nSearch TikTok videos, collect creator videos, and run product, trend, competitor, and content insights through Gecho Bridge MCP.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gecho-ai](https://clawhub.ai/user/gecho-ai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users, developers, and social media researchers use this skill to search TikTok videos, collect creator metadata, and start product, trend, competitor, or content insight jobs through Gecho Bridge MCP.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill depends on Gecho's MCP server, Chrome extension, and a logged-in TikTok browser session.\n\nMitigation: Install only if comfortable with that workflow, keep the extension logged in intentionally, and review Gecho extension and npm package provenance before use.\n\nRisk: Search and insight jobs can export result JSON files to a local save directory.\n\nMitigation: Use a workspace save directory and review exported files before sharing or committing them.\n\nRisk: TikTok login walls, CAPTCHA, frozen browser tabs, or missing MCP tools can prevent searches or insight jobs from running.\n\nMitigation: Confirm MCP configuration, Node/npx availability, Gecho extension login, and an active TikTok web session before starting a job.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gecho-ai/skills/tiktok-search)\n- [Gecho website](https://gecho.ai/)\n- [Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n- [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- [OpenClaw setup video](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- [Hermes setup video](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n\n## Skill Output:\n\n**Output Type(s):** [Markdown, Code, Shell commands, Configuration instructions, Guidance]\n\n**Output Format:** [Markdown with inline shell commands and structured result summaries]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May summarize TikTok result metadata, report saved JSON file paths, or return asynchronous insight job IDs.]\n\n## Skill Version(s):\n\n1.1.36 (source: server release metadata and artifact _meta.json)\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 v1.1.30: 3 files, 8235 bytes\n\nFiles: _meta.json (133b), skill-card.md (3501b), SKILL.md (19129b)\n\nFile v1.1.30:SKILL.md\n\n---\nname: tiktok-search\ndescription: Search TikTok videos, collect creator videos, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server.\nmetadata:\n  openclaw:\n    os: [\"darwin\", \"linux\", \"win32\"]\n    requires:\n      bins: [\"node\", \"npx\"]\n  hermes:\n    tags: [tiktok, search, insight, gecho, mcp]\n    category: social-media\n    os: [darwin, linux, windows]\n---\n\n# TikTok Search & Insight by Gecho\n\nSearch TikTok from an AI chat, collect structured video metadata, and run async product or trend insight jobs through the official Gecho Bridge MCP workflow.\n\nThis is the default TikTok aggregate Skill for Gecho. It covers TikTok video search, creator collection, and TikTok insight. Single-tool TikTok Skills may exist for distribution and search traffic, but this Skill is the recommended default for users who want the complete TikTok research workflow.\n\n## Critical prerequisite: read before use\n\nGecho Skills must be used with the Gecho Chrome extension. You must be logged in to both your Gecho account in the extension and TikTok in the Chrome web app. If either login is missing, all TikTok search and insight features may fail even though the Skill is installed.\n\n## 3-step quick start\n\n### Step 1: Install the Gecho Chrome extension\n\n1. Open the [Gecho Chrome extension download page](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb).\n2. Click `Add to Chrome`, then confirm `Add extension`.\n\n### Step 2: Log in to the Gecho extension\n\nOpen the Gecho extension in Chrome and log in to your Gecho account. Keep the extension online.\n\n### Step 3: Log in to TikTok web\n\nOpen TikTok in Chrome and log in to the TikTok web app. Keep the logged-in TikTok tab open.\n\nAfter setup is complete, return to the OpenClaw dashboard or Hermes and ask: \"Search xxx on TikTok\". For example: \"Search hamburgers on TikTok\".\n\n## Official links and setup help\n\n- Website: [gecho.ai](https://gecho.ai/)\n- GitHub: [gecho-ai/gecho-bridge](https://github.com/gecho-ai/gecho-bridge)\n- YouTube channel: [@Gecho-AI](https://www.youtube.com/@Gecho-AI)\n- Chrome extension: [Gecho Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- OpenClaw setup video: [OpenClaw + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- Hermes setup video: [Hermes + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n- Discord: [https://discord.gg/RFDVZMR6Tn](https://discord.gg/RFDVZMR6Tn)\n- WeCom group QR code: [qywx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg)\n- 1:1 support QR code: [wx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg)\n\n## What this skill does\n\n- Finds high-performing TikTok videos for a keyword.\n- Collects titles, authors, engagement data, and video links.\n- Saves full raw result sets to a local JSON file.\n- Starts async insight jobs for product research, competitor analysis, and trend discovery.\n\nBest-fit prompts:\n\n- \"Search TikTok for portable blender and show the top liked videos.\"\n- \"Find winning hooks for cat toy videos.\"\n- \"Run product opportunity insight for outdoor picnic mat.\"\n- \"Check the status of my previous TikTok insight job.\"\n\n## Important: Skill-only install is not enough\n\nThis Skill is the instruction layer. It tells the AI when and how to use Gecho.\n\nTo actually run TikTok searches, the user also needs:\n\n- the Gecho Bridge MCP server\n- the Gecho Chrome extension\n- Chrome with TikTok logged in\n- the Gecho extension logged in to a Gecho account and online\n\nIf the user installed only this Skill from ClawHub, searches will not work until the Gecho Bridge MCP server is configured. In that case, use the MCP setup path below.\n\nAlready installed Gecho Bridge? If `@gecho-ai/gecho-bridge-bundle` is installed and the Gecho MCP tools are visible, no extra MCP setup is needed for this Skill.\n\n## Quick start\n\n### OpenClaw Skill install: configure MCP\n\nIf this Skill is already installed in OpenClaw, configure the Gecho Bridge MCP server once:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\n### Optional: OpenClaw Bundle Plugin\n\nIf the user has not installed this Skill yet and prefers plugin management, the bundle plugin can install Gecho with the MCP entry configured:\n\n```bash\nopenclaw plugins install clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\nTo upgrade later:\n\n```bash\nopenclaw plugins update clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\n### Hermes setup\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\nIf Hermes cannot find `npx` even though Node.js is installed, use the absolute `npx` path. On many macOS Homebrew installs this is:\n\n```bash\nhermes mcp add gecho-bridge --command /opt/homebrew/bin/npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n## First-run checklist\n\nBefore the first search, make sure:\n\n- Node.js `>= 18` is available.\n- The [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) is installed.\n- Chrome is open with TikTok logged in.\n- The Gecho extension is logged in to a Gecho account and online.\n- The TikTok tab is not blocked by CAPTCHA, login walls, or a frozen page.\n\nFull setup guide:\n[Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n\nFor videos and support links, see the official links section above.\n\n## Tool choice\n\n| User goal | Use tool | Notes |\n|---|---|---|\n| Search TikTok videos and collect metadata | `tiktok_search` | Returns results directly |\n| Collect videos from a TikTok creator | `tiktok_influencer` | Uses the creator's `uniqueId` |\n| Analyze a niche, product, trend, or competitor opportunity | `tiktok_insight` | Starts an async job and returns a `jobId` |\n| Check an existing async insight job | `check_insight_status` | Use the `jobId` returned by `tiktok_insight` |\n\nUse this aggregate Skill for broad TikTok research requests. If the user explicitly asks for one exact raw tool only, such as \"run tiktok_search\", a matching single-tool distribution Skill may be used instead when available.\n\n## Official MCP tools\n\n### `tiktok_search`\n\nSearches TikTok for a keyword, scrolls the page through the Gecho browser extension, returns structured metadata, and saves the full result set.\n\nParameters:\n\n- `query` string, required: search keyword or phrase.\n- `targetCount` number, optional: requested result count; defaults to `100`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A JSON array of video metadata, plus a saved local file path when results are written successfully.\n\n### `tiktok_influencer`\n\nCollects videos published by one TikTok creator.\n\nParameters:\n\n- `uniqueId` string, required: the creator's TikTok `uniqueId` (for example, `zachking`).\n- `targetCount` number, optional: requested video count; defaults to `100`. Keep requests at or below `500`.\n\n### `tiktok_insight`\n\nStarts an asynchronous TikTok insight job for product, trend, or competitor research.\n\nParameters:\n\n- `query` string, required: search keyword or product/category phrase.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A `jobId`. The final result must be checked later with `check_insight_status`.\n\n### `check_insight_status`\n\nChecks the status or final result of an existing insight job.\n\nParameters:\n\n- `jobId` string, required: the job ID returned by `tiktok_insight`.\n\nExpected result:\n\n- `running`, `error`, or completed insight data.\n\n## Agent execution rules\n\nUse this Skill before calling any Gecho TikTok MCP tool when the user asks to search TikTok, find trending videos, analyze competitors, collect TikTok metadata, discover winning products, or research keyword trends.\n\nCore rules:\n\n- Use only the official Gecho MCP tools: `tiktok_search`, `tiktok_influencer`, `tiktok_insight`, and `check_insight_status`.\n- Do not replace Gecho with WebSearch, browser automation, terminal scrapers, mcporter, unofficial APIs, or hand-written TikTok scraping.\n- Do not start more than one Gecho scraping or insight job in the same conversational turn.\n- Do not run Gecho scraping jobs in parallel because the workflow depends on one live browser tab and extension session.\n- If a tool fails, times out, or returns an error, stop and report the exact failure reason.\n- If `tiktok_search` returns no items, do not rewrite, translate, broaden, or retry the keyword automatically.\n- If `tiktok_insight` starts successfully, report the `jobId` and explain that the user should check status later.\n- If `check_insight_status` says the job is still running, tell the user to wait before checking again.\n- If the official Gecho MCP tools are unavailable in the current session, provide setup instructions instead of probing the environment.\n- On first-run setup guidance, missing MCP tools, extension/session problems, timeouts, save failures, or any tool error, include the setup and support links block below.\n- Do not add the setup and support links block to normal successful search or insight responses unless the user asks for setup help.\n- Do not configure, edit, repair, or rewrite OpenClaw/Hermes/MCP settings on the user's behalf. Provide the setup commands and ask the user to run or approve them outside the tool workflow.\n- Do not use this Skill for TikTok Shop, X/Twitter, Amazon, or other platform workflows; those should have their own Gecho aggregate Skills.\n\nAllowed status behavior:\n\n- A user may explicitly ask to check one existing `jobId`; in that case, call `check_insight_status` once.\n- Do not start a new `tiktok_search` or `tiktok_insight` job in the same turn after checking status unless the user asks again in a later turn.\n\n## Standard workflows\n\n### Search workflow\n\n1. Use the exact keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_search`.\n4. If the result is empty, say that the exact keyword returned no results and stop.\n5. If results are present, summarize only the top 3 to 5 items and provide the saved file path.\n\n### Insight workflow\n\n1. Use the exact product, niche, or keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_insight`.\n4. Report the returned `jobId`.\n5. Tell the user the job may take several minutes and should be checked later with `check_insight_status`.\n\n### Status workflow\n\n1. Call `check_insight_status` with the provided `jobId`.\n2. If still running, report the current status and recommend waiting about 60 seconds.\n3. If completed, summarize the key findings and provide the saved file path.\n4. If failed, report the exact error and stop.\n\n## Setup and support links block\n\nUse this compact block on first-run guidance and all setup or failure responses. Keep it near the end of the answer, after the exact fix.\n\n````markdown\nHelpful Gecho links:\n\n- Website: https://gecho.ai/\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Support: Discord https://discord.gg/RFDVZMR6Tn, WeCom group QR https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg, 1:1 support QR https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n````\n\n## Setup-missing response\n\nUse this when the official Gecho MCP tools are not available, or when the user only installed the Skill and cannot run searches yet.\n\nUse the template below as the canonical setup-missing response. The agent may translate the prose to match the user's language, but must not summarize it into a shorter checklist unless the user explicitly asks for a short version.\n\nRequired content that must not be omitted:\n\n- State that Gecho Bridge MCP is not configured and the Skill alone cannot run TikTok.\n- Include all 3 requirements: MCP configured, Gecho Chrome extension installed and logged in, TikTok web logged in and tab kept open.\n- Include the OpenClaw MCP setup command and `openclaw mcp list` verification command.\n- Include the Chrome extension link.\n- Include after-setup wording that tells the user to return to OpenClaw Dashboard or Hermes and retry the request.\n- Include all related links: website, OpenClaw setup video, Hermes setup video, YouTube channel, GitHub/README, Discord, WeCom group QR code, and 1:1 support QR code.\n- Include related Gecho Skills listed in the template.\n- Do not recommend `openclaw plugins install` in this setup-missing response.\n\n````markdown\nGecho Bridge is not ready yet.\n\nThis Skill is installed, but the official Gecho Bridge MCP tools are not available in this session yet. Installing the Skill alone does not start the TikTok search service.\n\nGecho requires all 3 items below before TikTok search or insight can run:\n\n1. Gecho Bridge MCP is configured.\n2. The Gecho Chrome extension is installed and logged in to a Gecho account.\n3. TikTok web is logged in inside Chrome, with the TikTok tab kept open.\n\nFollow these first-time setup steps:\n\n**1. Install the Gecho Chrome extension**\n\nChrome Web Store:\nhttps://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb\n\nAfter installing, open the extension and log in to your Gecho account.\n\n**2. Configure Gecho Bridge MCP**\n\nOpenClaw MCP setup:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\nFor Hermes:\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n**3. Log in to TikTok web**\n\nOpen TikTok in Chrome, log in, and keep the logged-in TikTok tab open.\n\nAfter setup, return to OpenClaw Dashboard or Hermes and ask again, for example:\n\"Search computers on TikTok\"\n\n\n**Related links**\n\n- Website: https://gecho.ai/\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Discord support: https://discord.gg/RFDVZMR6Tn\n- WeCom group QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg\n- 1:1 support QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n\n**Related Gecho Skills**\n\n- `tiktok-video-search`: TikTok video search and metadata collection.\n- `tiktok-insight`: TikTok product, trend, competitor, and content insight jobs.\n- `tiktok-search`: complete TikTok search and insight workflow.\n````\n\n## Troubleshooting\n\n| Situation | What to do |\n|---|---|\n| MCP tools are missing | Give the setup-missing response. Do not run local shell probes. |\n| User installed only the Skill | Explain that Skill-only install is not enough and provide the OpenClaw MCP setup command. |\n| Hermes MCP tools are missing | Provide the `hermes mcp add ...` command. Do not inspect or rewrite Hermes config files. |\n| Extension not connected | Ask the user to enable/login to the Gecho Chrome extension and keep a logged-in TikTok tab open. |\n| CAPTCHA or login wall | Ask the user to resolve it manually in Chrome, then retry in a later turn. |\n| Request timeout | Report the timeout, mention stuck page/network/CAPTCHA as likely causes, and stop. |\n| Empty search results | Say the exact keyword returned no results and ask the user to choose another keyword manually. |\n| Insight still running | Report running status and recommend checking again after about 60 seconds. |\n| Failed to save results | Ask the user to provide a valid absolute directory path with write permission. |\n\n## FAQ\n\n### Why is the Chrome extension required? Can't I just use the web page?\n\nGecho needs real-time platform data from a live browser session, such as TikTok videos and other platform data in Gecho workflows. The Chrome extension connects the AI workflow to the user's logged-in Chrome session; the Skill page alone cannot collect this data.\n\n### Why do I need to log in to TikTok? Can I use it without login?\n\nTikTok limits content access for logged-out users. After you log in, the extension can access the complete data available in your browser session, such as video captions/scripts, comments, engagement data, and other signals when available.\n\nGecho does not ask for or collect your TikTok password, private account information, or publish anything on your behalf.\n\n### Need help?\n\nJoin the [WeCom group](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg) for community support, or scan the [1:1 support QR code](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg) for personal help.\n\n## Output guidelines\n\nFor successful search:\n\n- Say the search completed.\n- Include total result count if available.\n- Include saved file path if available.\n- Show top 3 to 5 results only.\n- Do not paste the full raw JSON into chat.\n\nFor successful insight start:\n\n- Say the insight job started.\n- Include the `jobId`.\n- Include expected saved path if available.\n- Tell the user to check status later.\n\nFor completed insight:\n\n- Summarize key findings.\n- Include saved file path.\n- Avoid claiming conclusions not supported by returned tool data.\n\nFor failures:\n\n- Report the exact tool error or failure state.\n- Provide only the relevant fix from Troubleshooting.\n- Include the setup and support links block so the user can continue setup through docs, videos, or support.\n- Do not retry in the same turn.\n\n## Scope and limits\n\nThis Skill should:\n\n- Help users complete the official Gecho setup when prerequisites are missing.\n- Route TikTok search and insight requests to the official Gecho MCP tools.\n- Keep search, insight, and status-check flows explicit.\n- Summarize results without flooding the chat.\n\nThis Skill must never:\n\n- Pretend the Skill page alone is enough when MCP is missing.\n- Pretend `tiktok_insight` is synchronous.\n- Use unofficial TikTok scraping workflows.\n- Invent results when the tool returns no data.\n- Solve CAPTCHA, log in to TikTok, or operate the user's browser outside the official Gecho MCP workflow.\n\nFile v1.1.30:_meta.json\n\n{\n  \"ownerId\": \"kn7b732kkfejrjfkjq2h0hc9qn84nv5k\",\n  \"slug\": \"tiktok-search\",\n  \"version\": \"1.1.30\",\n  \"publishedAt\": 1785571851775\n}\n\nFile v1.1.30:skill-card.md\n\n## Description: <br>\nSearch TikTok videos, collect creator videos, and run product, trend, competitor, and content insights through Gecho Bridge MCP, requiring the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[gecho-ai](https://clawhub.ai/user/gecho-ai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nExternal users and developers use this skill to search TikTok, collect structured creator and video metadata, and start product, trend, competitor, or content insight jobs through Gecho Bridge. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: TikTok workflows will not run unless Gecho Bridge MCP, the Gecho Chrome extension, a logged-in Gecho account, and a logged-in TikTok web session are configured. <br>\nMitigation: Complete the documented setup steps and verify that the official Gecho MCP tools are available before attempting a search or insight job. <br>\nRisk: The workflow uses a live browser session and Chrome extension to collect TikTok metadata. <br>\nMitigation: Review the external MCP package and extension permissions, and avoid using accounts or browser sessions that contain sensitive private activity unless the integration is trusted. <br>\nRisk: Result saving can fail or write to an unintended location when the save directory is unreliable. <br>\nMitigation: Use a known absolute directory with appropriate write permissions, or omit the save directory and let Gecho use its default data location. <br>\nRisk: Login walls, CAPTCHA, frozen pages, network issues, or TikTok session problems can block collection. <br>\nMitigation: Resolve browser-side blocks manually, stop on tool errors or timeouts, and report the exact failure instead of retrying automatically. <br>\nRisk: TikTok insight jobs are asynchronous and may not complete in the same turn. <br>\nMitigation: Return the job ID, tell the user the job can take several minutes, and check status later with the official status tool. <br>\n\n\n## Reference(s): <br>\n- [ClawHub skill page](https://clawhub.ai/gecho-ai/skills/tiktok-search) <br>\n- [Gecho website](https://gecho.ai/) <br>\n- [Gecho Bridge repository](https://github.com/gecho-ai/gecho-bridge) <br>\n- [Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md) <br>\n- [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) <br>\n- [OpenClaw setup video](https://www.youtube.com/watch?v=ggwY9hISHcQ) <br>\n- [Hermes setup video](https://www.youtube.com/watch?v=zHKnuWnxt_c) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, shell commands, configuration, JSON, guidance] <br>\n**Output Format:** [Markdown guidance with shell commands and summarized JSON-like MCP/tool results] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May include saved local file paths and summarized TikTok metadata; full raw result sets are saved to files rather than pasted.] <br>\n\n## Skill Version(s): <br>\n1.1.30 (source: server release metadata and artifact _meta.json) <br>\n\n## Ethical Considerations: <br>\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. <br>\n\nArchive v1.1.29: 3 files, 7937 bytes\n\nFiles: _meta.json (133b), skill-card.md (2719b), SKILL.md (19129b)\n\nFile v1.1.29:SKILL.md\n\n---\nname: tiktok-search\ndescription: Search TikTok videos, collect creator videos, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server.\nmetadata:\n  openclaw:\n    os: [\"darwin\", \"linux\", \"win32\"]\n    requires:\n      bins: [\"node\", \"npx\"]\n  hermes:\n    tags: [tiktok, search, insight, gecho, mcp]\n    category: social-media\n    os: [darwin, linux, windows]\n---\n\n# TikTok Search & Insight by Gecho\n\nSearch TikTok from an AI chat, collect structured video metadata, and run async product or trend insight jobs through the official Gecho Bridge MCP workflow.\n\nThis is the default TikTok aggregate Skill for Gecho. It covers TikTok video search, creator collection, and TikTok insight. Single-tool TikTok Skills may exist for distribution and search traffic, but this Skill is the recommended default for users who want the complete TikTok research workflow.\n\n## Critical prerequisite: read before use\n\nGecho Skills must be used with the Gecho Chrome extension. You must be logged in to both your Gecho account in the extension and TikTok in the Chrome web app. If either login is missing, all TikTok search and insight features may fail even though the Skill is installed.\n\n## 3-step quick start\n\n### Step 1: Install the Gecho Chrome extension\n\n1. Open the [Gecho Chrome extension download page](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb).\n2. Click `Add to Chrome`, then confirm `Add extension`.\n\n### Step 2: Log in to the Gecho extension\n\nOpen the Gecho extension in Chrome and log in to your Gecho account. Keep the extension online.\n\n### Step 3: Log in to TikTok web\n\nOpen TikTok in Chrome and log in to the TikTok web app. Keep the logged-in TikTok tab open.\n\nAfter setup is complete, return to the OpenClaw dashboard or Hermes and ask: \"Search xxx on TikTok\". For example: \"Search hamburgers on TikTok\".\n\n## Official links and setup help\n\n- Website: [gecho.ai](https://gecho.ai/)\n- GitHub: [gecho-ai/gecho-bridge](https://github.com/gecho-ai/gecho-bridge)\n- YouTube channel: [@Gecho-AI](https://www.youtube.com/@Gecho-AI)\n- Chrome extension: [Gecho Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- OpenClaw setup video: [OpenClaw + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- Hermes setup video: [Hermes + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n- Discord: [https://discord.gg/RFDVZMR6Tn](https://discord.gg/RFDVZMR6Tn)\n- WeCom group QR code: [qywx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg)\n- 1:1 support QR code: [wx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg)\n\n## What this skill does\n\n- Finds high-performing TikTok videos for a keyword.\n- Collects titles, authors, engagement data, and video links.\n- Saves full raw result sets to a local JSON file.\n- Starts async insight jobs for product research, competitor analysis, and trend discovery.\n\nBest-fit prompts:\n\n- \"Search TikTok for portable blender and show the top liked videos.\"\n- \"Find winning hooks for cat toy videos.\"\n- \"Run product opportunity insight for outdoor picnic mat.\"\n- \"Check the status of my previous TikTok insight job.\"\n\n## Important: Skill-only install is not enough\n\nThis Skill is the instruction layer. It tells the AI when and how to use Gecho.\n\nTo actually run TikTok searches, the user also needs:\n\n- the Gecho Bridge MCP server\n- the Gecho Chrome extension\n- Chrome with TikTok logged in\n- the Gecho extension logged in to a Gecho account and online\n\nIf the user installed only this Skill from ClawHub, searches will not work until the Gecho Bridge MCP server is configured. In that case, use the MCP setup path below.\n\nAlready installed Gecho Bridge? If `@gecho-ai/gecho-bridge-bundle` is installed and the Gecho MCP tools are visible, no extra MCP setup is needed for this Skill.\n\n## Quick start\n\n### OpenClaw Skill install: configure MCP\n\nIf this Skill is already installed in OpenClaw, configure the Gecho Bridge MCP server once:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\n### Optional: OpenClaw Bundle Plugin\n\nIf the user has not installed this Skill yet and prefers plugin management, the bundle plugin can install Gecho with the MCP entry configured:\n\n```bash\nopenclaw plugins install clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\nTo upgrade later:\n\n```bash\nopenclaw plugins update clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\n### Hermes setup\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\nIf Hermes cannot find `npx` even though Node.js is installed, use the absolute `npx` path. On many macOS Homebrew installs this is:\n\n```bash\nhermes mcp add gecho-bridge --command /opt/homebrew/bin/npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n## First-run checklist\n\nBefore the first search, make sure:\n\n- Node.js `>= 18` is available.\n- The [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) is installed.\n- Chrome is open with TikTok logged in.\n- The Gecho extension is logged in to a Gecho account and online.\n- The TikTok tab is not blocked by CAPTCHA, login walls, or a frozen page.\n\nFull setup guide:\n[Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n\nFor videos and support links, see the official links section above.\n\n## Tool choice\n\n| User goal | Use tool | Notes |\n|---|---|---|\n| Search TikTok videos and collect metadata | `tiktok_search` | Returns results directly |\n| Collect videos from a TikTok creator | `tiktok_influencer` | Uses the creator's `uniqueId` |\n| Analyze a niche, product, trend, or competitor opportunity | `tiktok_insight` | Starts an async job and returns a `jobId` |\n| Check an existing async insight job | `check_insight_status` | Use the `jobId` returned by `tiktok_insight` |\n\nUse this aggregate Skill for broad TikTok research requests. If the user explicitly asks for one exact raw tool only, such as \"run tiktok_search\", a matching single-tool distribution Skill may be used instead when available.\n\n## Official MCP tools\n\n### `tiktok_search`\n\nSearches TikTok for a keyword, scrolls the page through the Gecho browser extension, returns structured metadata, and saves the full result set.\n\nParameters:\n\n- `query` string, required: search keyword or phrase.\n- `targetCount` number, optional: requested result count; defaults to `100`.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A JSON array of video metadata, plus a saved local file path when results are written successfully.\n\n### `tiktok_influencer`\n\nCollects videos published by one TikTok creator.\n\nParameters:\n\n- `uniqueId` string, required: the creator's TikTok `uniqueId` (for example, `zachking`).\n- `targetCount` number, optional: requested video count; defaults to `100`. Keep requests at or below `500`.\n\n### `tiktok_insight`\n\nStarts an asynchronous TikTok insight job for product, trend, or competitor research.\n\nParameters:\n\n- `query` string, required: search keyword or product/category phrase.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A `jobId`. The final result must be checked later with `check_insight_status`.\n\n### `check_insight_status`\n\nChecks the status or final result of an existing insight job.\n\nParameters:\n\n- `jobId` string, required: the job ID returned by `tiktok_insight`.\n\nExpected result:\n\n- `running`, `error`, or completed insight data.\n\n## Agent execution rules\n\nUse this Skill before calling any Gecho TikTok MCP tool when the user asks to search TikTok, find trending videos, analyze competitors, collect TikTok metadata, discover winning products, or research keyword trends.\n\nCore rules:\n\n- Use only the official Gecho MCP tools: `tiktok_search`, `tiktok_influencer`, `tiktok_insight`, and `check_insight_status`.\n- Do not replace Gecho with WebSearch, browser automation, terminal scrapers, mcporter, unofficial APIs, or hand-written TikTok scraping.\n- Do not start more than one Gecho scraping or insight job in the same conversational turn.\n- Do not run Gecho scraping jobs in parallel because the workflow depends on one live browser tab and extension session.\n- If a tool fails, times out, or returns an error, stop and report the exact failure reason.\n- If `tiktok_search` returns no items, do not rewrite, translate, broaden, or retry the keyword automatically.\n- If `tiktok_insight` starts successfully, report the `jobId` and explain that the user should check status later.\n- If `check_insight_status` says the job is still running, tell the user to wait before checking again.\n- If the official Gecho MCP tools are unavailable in the current session, provide setup instructions instead of probing the environment.\n- On first-run setup guidance, missing MCP tools, extension/session problems, timeouts, save failures, or any tool error, include the setup and support links block below.\n- Do not add the setup and support links block to normal successful search or insight responses unless the user asks for setup help.\n- Do not configure, edit, repair, or rewrite OpenClaw/Hermes/MCP settings on the user's behalf. Provide the setup commands and ask the user to run or approve them outside the tool workflow.\n- Do not use this Skill for TikTok Shop, X/Twitter, Amazon, or other platform workflows; those should have their own Gecho aggregate Skills.\n\nAllowed status behavior:\n\n- A user may explicitly ask to check one existing `jobId`; in that case, call `check_insight_status` once.\n- Do not start a new `tiktok_search` or `tiktok_insight` job in the same turn after checking status unless the user asks again in a later turn.\n\n## Standard workflows\n\n### Search workflow\n\n1. Use the exact keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_search`.\n4. If the result is empty, say that the exact keyword returned no results and stop.\n5. If results are present, summarize only the top 3 to 5 items and provide the saved file path.\n\n### Insight workflow\n\n1. Use the exact product, niche, or keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_insight`.\n4. Report the returned `jobId`.\n5. Tell the user the job may take several minutes and should be checked later with `check_insight_status`.\n\n### Status workflow\n\n1. Call `check_insight_status` with the provided `jobId`.\n2. If still running, report the current status and recommend waiting about 60 seconds.\n3. If completed, summarize the key findings and provide the saved file path.\n4. If failed, report the exact error and stop.\n\n## Setup and support links block\n\nUse this compact block on first-run guidance and all setup or failure responses. Keep it near the end of the answer, after the exact fix.\n\n````markdown\nHelpful Gecho links:\n\n- Website: https://gecho.ai/\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Support: Discord https://discord.gg/RFDVZMR6Tn, WeCom group QR https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg, 1:1 support QR https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n````\n\n## Setup-missing response\n\nUse this when the official Gecho MCP tools are not available, or when the user only installed the Skill and cannot run searches yet.\n\nUse the template below as the canonical setup-missing response. The agent may translate the prose to match the user's language, but must not summarize it into a shorter checklist unless the user explicitly asks for a short version.\n\nRequired content that must not be omitted:\n\n- State that Gecho Bridge MCP is not configured and the Skill alone cannot run TikTok.\n- Include all 3 requirements: MCP configured, Gecho Chrome extension installed and logged in, TikTok web logged in and tab kept open.\n- Include the OpenClaw MCP setup command and `openclaw mcp list` verification command.\n- Include the Chrome extension link.\n- Include after-setup wording that tells the user to return to OpenClaw Dashboard or Hermes and retry the request.\n- Include all related links: website, OpenClaw setup video, Hermes setup video, YouTube channel, GitHub/README, Discord, WeCom group QR code, and 1:1 support QR code.\n- Include related Gecho Skills listed in the template.\n- Do not recommend `openclaw plugins install` in this setup-missing response.\n\n````markdown\nGecho Bridge is not ready yet.\n\nThis Skill is installed, but the official Gecho Bridge MCP tools are not available in this session yet. Installing the Skill alone does not start the TikTok search service.\n\nGecho requires all 3 items below before TikTok search or insight can run:\n\n1. Gecho Bridge MCP is configured.\n2. The Gecho Chrome extension is installed and logged in to a Gecho account.\n3. TikTok web is logged in inside Chrome, with the TikTok tab kept open.\n\nFollow these first-time setup steps:\n\n**1. Install the Gecho Chrome extension**\n\nChrome Web Store:\nhttps://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb\n\nAfter installing, open the extension and log in to your Gecho account.\n\n**2. Configure Gecho Bridge MCP**\n\nOpenClaw MCP setup:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\nFor Hermes:\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n**3. Log in to TikTok web**\n\nOpen TikTok in Chrome, log in, and keep the logged-in TikTok tab open.\n\nAfter setup, return to OpenClaw Dashboard or Hermes and ask again, for example:\n\"Search computers on TikTok\"\n\n\n**Related links**\n\n- Website: https://gecho.ai/\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Discord support: https://discord.gg/RFDVZMR6Tn\n- WeCom group QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg\n- 1:1 support QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg\n\n**Related Gecho Skills**\n\n- `tiktok-video-search`: TikTok video search and metadata collection.\n- `tiktok-insight`: TikTok product, trend, competitor, and content insight jobs.\n- `tiktok-search`: complete TikTok search and insight workflow.\n````\n\n## Troubleshooting\n\n| Situation | What to do |\n|---|---|\n| MCP tools are missing | Give the setup-missing response. Do not run local shell probes. |\n| User installed only the Skill | Explain that Skill-only install is not enough and provide the OpenClaw MCP setup command. |\n| Hermes MCP tools are missing | Provide the `hermes mcp add ...` command. Do not inspect or rewrite Hermes config files. |\n| Extension not connected | Ask the user to enable/login to the Gecho Chrome extension and keep a logged-in TikTok tab open. |\n| CAPTCHA or login wall | Ask the user to resolve it manually in Chrome, then retry in a later turn. |\n| Request timeout | Report the timeout, mention stuck page/network/CAPTCHA as likely causes, and stop. |\n| Empty search results | Say the exact keyword returned no results and ask the user to choose another keyword manually. |\n| Insight still running | Report running status and recommend checking again after about 60 seconds. |\n| Failed to save results | Ask the user to provide a valid absolute directory path with write permission. |\n\n## FAQ\n\n### Why is the Chrome extension required? Can't I just use the web page?\n\nGecho needs real-time platform data from a live browser session, such as TikTok videos and other platform data in Gecho workflows. The Chrome extension connects the AI workflow to the user's logged-in Chrome session; the Skill page alone cannot collect this data.\n\n### Why do I need to log in to TikTok? Can I use it without login?\n\nTikTok limits content access for logged-out users. After you log in, the extension can access the complete data available in your browser session, such as video captions/scripts, comments, engagement data, and other signals when available.\n\nGecho does not ask for or collect your TikTok password, private account information, or publish anything on your behalf.\n\n### Need help?\n\nJoin the [WeCom group](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg) for community support, or scan the [1:1 support QR code](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg) for personal help.\n\n## Output guidelines\n\nFor successful search:\n\n- Say the search completed.\n- Include total result count if available.\n- Include saved file path if available.\n- Show top 3 to 5 results only.\n- Do not paste the full raw JSON into chat.\n\nFor successful insight start:\n\n- Say the insight job started.\n- Include the `jobId`.\n- Include expected saved path if available.\n- Tell the user to check status later.\n\nFor completed insight:\n\n- Summarize key findings.\n- Include saved file path.\n- Avoid claiming conclusions not supported by returned tool data.\n\nFor failures:\n\n- Report the exact tool error or failure state.\n- Provide only the relevant fix from Troubleshooting.\n- Include the setup and support links block so the user can continue setup through docs, videos, or support.\n- Do not retry in the same turn.\n\n## Scope and limits\n\nThis Skill should:\n\n- Help users complete the official Gecho setup when prerequisites are missing.\n- Route TikTok search and insight requests to the official Gecho MCP tools.\n- Keep search, insight, and status-check flows explicit.\n- Summarize results without flooding the chat.\n\nThis Skill must never:\n\n- Pretend the Skill page alone is enough when MCP is missing.\n- Pretend `tiktok_insight` is synchronous.\n- Use unofficial TikTok scraping workflows.\n- Invent results when the tool returns no data.\n- Solve CAPTCHA, log in to TikTok, or operate the user's browser outside the official Gecho MCP workflow.\n\nFile v1.1.29:_meta.json\n\n{\n  \"ownerId\": \"kn7b732kkfejrjfkjq2h0hc9qn84nv5k\",\n  \"slug\": \"tiktok-search\",\n  \"version\": \"1.1.29\",\n  \"publishedAt\": 1784626118812\n}\n\nFile v1.1.29:skill-card.md\n\n## Description: <br>\nSearch TikTok videos, collect creator videos, and run product, trend, competitor, and content insights through Gecho Bridge MCP. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[gecho-ai](https://clawhub.ai/user/gecho-ai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nExternal users and developers use this skill to search TikTok, collect creator and video metadata, and run product, trend, competitor, or content insight workflows through Gecho Bridge MCP. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: The workflow depends on a third-party MCP server and Chrome extension connected to a logged-in TikTok browser session. <br>\nMitigation: Use it only when comfortable with Gecho accessing TikTok data visible in that browser session, and keep the extension and TikTok session limited to the intended research account. <br>\nRisk: The skill saves raw TikTok research results to local JSON files. <br>\nMitigation: Choose a trusted save directory and review saved files before sharing or retaining them. <br>\nRisk: TikTok search or insight features may fail if the MCP server, Chrome extension, Gecho login, or TikTok login is missing. <br>\nMitigation: Complete the documented Gecho Bridge setup, keep the extension online, and keep a logged-in TikTok tab open before running searches. <br>\n\n\n## Reference(s): <br>\n- [Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md) <br>\n- [Gecho Chrome Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) <br>\n- [Gecho Website](https://gecho.ai/) <br>\n- [OpenClaw Setup Video](https://www.youtube.com/watch?v=ggwY9hISHcQ) <br>\n- [Hermes Setup Video](https://www.youtube.com/watch?v=zHKnuWnxt_c) <br>\n- [Gecho Support Discord](https://discord.gg/RFDVZMR6Tn) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Text, Markdown, Shell commands, Configuration instructions, API calls, Files] <br>\n**Output Format:** [Markdown with setup commands, JSON result summaries, job IDs, and local file paths.] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [Successful searches summarize the top 3 to 5 items; insight jobs return a jobId and require a later status check.] <br>\n\n## Skill Version(s): <br>\n1.1.29 (source: release evidence and artifact metadata) <br>\n\n## Ethical Considerations: <br>\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. <br>\n\nArchive v1.1.24: 3 files, 7746 bytes\n\nFiles: _meta.json (133b), skill-card.md (2783b), SKILL.md (18339b)\n\nFile v1.1.24:SKILL.md\n\n---\nname: tiktok-search\ndescription: Search TikTok videos and run product, trend, competitor, and content insights through Gecho Bridge MCP. Use for TikTok keyword research, trending video discovery, winning-product analysis, and checking async insight jobs. Requires the Gecho Chrome extension, an active TikTok session, and the shared Gecho Bridge MCP server.\nmetadata:\n  openclaw:\n    os: [\"darwin\", \"linux\", \"win32\"]\n    requires:\n      bins: [\"node\", \"npx\"]\n  hermes:\n    tags: [tiktok, search, insight, gecho, mcp]\n    category: social-media\n    os: [darwin, linux, windows]\n---\n\n# TikTok Search & Insight by Gecho\n\nSearch TikTok from an AI chat, collect structured video metadata, and run async product or trend insight jobs through the official Gecho Bridge MCP workflow.\n\nThis is the default TikTok aggregate Skill for Gecho. It covers the TikTok video search and TikTok insight workflow. Single-tool TikTok Skills may exist for distribution and search traffic, but this Skill is the recommended default for users who want the complete TikTok research workflow.\n\n## Critical prerequisite: read before use\n\nGecho Skills must be used with the Gecho Chrome extension. You must be logged in to both your Gecho account in the extension and TikTok in the Chrome web app. If either login is missing, all TikTok search and insight features may fail even though the Skill is installed.\n\n## 3-step quick start\n\n### Step 1: Install the Gecho Chrome extension\n\n1. Open the [Gecho Chrome extension download page](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb).\n2. Click `Add to Chrome`, then confirm `Add extension`.\n\n### Step 2: Log in to the Gecho extension\n\nOpen the Gecho extension in Chrome and log in to your Gecho account. Keep the extension online.\n\n### Step 3: Log in to TikTok web\n\nOpen TikTok in Chrome and log in to the TikTok web app. Keep the logged-in TikTok tab open.\n\nAfter setup is complete, return to the OpenClaw dashboard or Hermes and ask: \"Search xxx on TikTok\". For example: \"Search hamburgers on TikTok\".\n\n## Official links and setup help\n\n- Website: [gecho.ai](https://gecho.ai/)\n- GitHub: [gecho-ai/gecho-bridge](https://github.com/gecho-ai/gecho-bridge)\n- YouTube channel: [@Gecho-AI](https://www.youtube.com/@Gecho-AI)\n- Chrome extension: [Gecho Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- OpenClaw setup video: [OpenClaw + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- Hermes setup video: [Hermes + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n- Discord: [https://discord.gg/RFDVZMR6Tn](https://discord.gg/RFDVZMR6Tn)\n- WeCom group QR code: [qywx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg)\n\n## What this skill does\n\n- Finds high-performing TikTok videos for a keyword.\n- Collects titles, authors, engagement data, and video links.\n- Saves full raw result sets to a local JSON file.\n- Starts async insight jobs for product research, competitor analysis, and trend discovery.\n\nBest-fit prompts:\n\n- \"Search TikTok for portable blender and show the top liked videos.\"\n- \"Find winning hooks for cat toy videos.\"\n- \"Run product opportunity insight for outdoor picnic mat.\"\n- \"Check the status of my previous TikTok insight job.\"\n\n## Important: Skill-only install is not enough\n\nThis Skill is the instruction layer. It tells the AI when and how to use Gecho.\n\nTo actually run TikTok searches, the user also needs:\n\n- the Gecho Bridge MCP server\n- the Gecho Chrome extension\n- Chrome with TikTok logged in\n- the Gecho extension logged in to a Gecho account and online\n\nIf the user installed only this Skill from ClawHub, searches will not work until the Gecho Bridge MCP server is configured. In that case, use the MCP setup path below.\n\nAlready installed Gecho Bridge? If `@gecho-ai/gecho-bridge-bundle` is installed and the Gecho MCP tools are visible, no extra MCP setup is needed for this Skill.\n\n## Quick start\n\n### OpenClaw Skill install: configure MCP\n\nIf this Skill is already installed in OpenClaw, configure the Gecho Bridge MCP server once:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\n### Optional: OpenClaw Bundle Plugin\n\nIf the user has not installed this Skill yet and prefers plugin management, the bundle plugin can install Gecho with the MCP entry configured:\n\n```bash\nopenclaw plugins install clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\nTo upgrade later:\n\n```bash\nopenclaw plugins update clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart\n```\n\n### Hermes setup\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\nIf Hermes cannot find `npx` even though Node.js is installed, use the absolute `npx` path. On many macOS Homebrew installs this is:\n\n```bash\nhermes mcp add gecho-bridge --command /opt/homebrew/bin/npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n## First-run checklist\n\nBefore the first search, make sure:\n\n- Node.js `>= 18` is available.\n- The [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) is installed.\n- Chrome is open with TikTok logged in.\n- The Gecho extension is logged in to a Gecho account and online.\n- The TikTok tab is not blocked by CAPTCHA, login walls, or a frozen page.\n\nFull setup guide:\n[Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n\nFor videos and support links, see the official links section above.\n\n## Tool choice\n\n| User goal | Use tool | Notes |\n|---|---|---|\n| Search TikTok videos and collect metadata | `tiktok_search` | Returns results directly |\n| Analyze a niche, product, trend, or competitor opportunity | `tiktok_insight` | Starts an async job and returns a `jobId` |\n| Check an existing async insight job | `check_insight_status` | Use the `jobId` returned by `tiktok_insight` |\n\nUse this aggregate Skill for broad TikTok research requests. If the user explicitly asks for one exact raw tool only, such as \"run tiktok_search\", a matching single-tool distribution Skill may be used instead when available.\n\n## Official MCP tools\n\n### `tiktok_search`\n\nSearches TikTok for a keyword, scrolls the page through the Gecho browser extension, returns structured metadata, and saves the full result set.\n\nParameters:\n\n- `query` string, required: search keyword or phrase.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A JSON array of video metadata, plus a saved local file path when results are written successfully.\n\n### `tiktok_insight`\n\nStarts an asynchronous TikTok insight job for product, trend, or competitor research.\n\nParameters:\n\n- `query` string, required: search keyword or product/category phrase.\n- `save_dir` string, optional: absolute directory path for saving results. Do not pass a `.json` filename. Omit this parameter if no reliable absolute directory is available.\n\nExpected result:\n\n- A `jobId`. The final result must be checked later with `check_insight_status`.\n\n### `check_insight_status`\n\nChecks the status or final result of an existing insight job.\n\nParameters:\n\n- `jobId` string, required: the job ID returned by `tiktok_insight`.\n\nExpected result:\n\n- `running`, `error`, or completed insight data.\n\n## Agent execution rules\n\nUse this Skill before calling any Gecho TikTok MCP tool when the user asks to search TikTok, find trending videos, analyze competitors, collect TikTok metadata, discover winning products, or research keyword trends.\n\nCore rules:\n\n- Use only the official Gecho MCP tools: `tiktok_search`, `tiktok_insight`, and `check_insight_status`.\n- Do not replace Gecho with WebSearch, browser automation, terminal scrapers, mcporter, unofficial APIs, or hand-written TikTok scraping.\n- Do not start more than one Gecho scraping or insight job in the same conversational turn.\n- Do not run Gecho scraping jobs in parallel because the workflow depends on one live browser tab and extension session.\n- If a tool fails, times out, or returns an error, stop and report the exact failure reason.\n- If `tiktok_search` returns no items, do not rewrite, translate, broaden, or retry the keyword automatically.\n- If `tiktok_insight` starts successfully, report the `jobId` and explain that the user should check status later.\n- If `check_insight_status` says the job is still running, tell the user to wait before checking again.\n- If the official Gecho MCP tools are unavailable in the current session, provide setup instructions instead of probing the environment.\n- On first-run setup guidance, missing MCP tools, extension/session problems, timeouts, save failures, or any tool error, include the setup and support links block below.\n- Do not add the setup and support links block to normal successful search or insight responses unless the user asks for setup help.\n- Do not configure, edit, repair, or rewrite OpenClaw/Hermes/MCP settings on the user's behalf. Provide the setup commands and ask the user to run or approve them outside the tool workflow.\n- Do not use this Skill for TikTok Shop, X/Twitter, Amazon, or other platform workflows; those should have their own Gecho aggregate Skills.\n\nAllowed status behavior:\n\n- A user may explicitly ask to check one existing `jobId`; in that case, call `check_insight_status` once.\n- Do not start a new `tiktok_search` or `tiktok_insight` job in the same turn after checking status unless the user asks again in a later turn.\n\n## Standard workflows\n\n### Search workflow\n\n1. Use the exact keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_search`.\n4. If the result is empty, say that the exact keyword returned no results and stop.\n5. If results are present, summarize only the top 3 to 5 items and provide the saved file path.\n\n### Insight workflow\n\n1. Use the exact product, niche, or keyword requested by the user.\n2. If the user did not provide `save_dir`, choose a safe absolute directory path in the current workspace. If no reliable absolute directory is available, omit `save_dir` and let Gecho use its default data directory.\n3. Call `tiktok_insight`.\n4. Report the returned `jobId`.\n5. Tell the user the job may take several minutes and should be checked later with `check_insight_status`.\n\n### Status workflow\n\n1. Call `check_insight_status` with the provided `jobId`.\n2. If still running, report the current status and recommend waiting about 60 seconds.\n3. If completed, summarize the key findings and provide the saved file path.\n4. If failed, report the exact error and stop.\n\n## Setup and support links block\n\nUse this compact block on first-run guidance and all setup or failure responses. Keep it near the end of the answer, after the exact fix.\n\n````markdown\nHelpful Gecho links:\n\n- Website: https://gecho.ai/\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Support: Discord https://discord.gg/RFDVZMR6Tn, WeCom QR https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg\n````\n\n## Setup-missing response\n\nUse this when the official Gecho MCP tools are not available, or when the user only installed the Skill and cannot run searches yet.\n\nUse the template below as the canonical setup-missing response. The agent may translate the prose to match the user's language, but must not summarize it into a shorter checklist unless the user explicitly asks for a short version.\n\nRequired content that must not be omitted:\n\n- State that Gecho Bridge MCP is not configured and the Skill alone cannot run TikTok.\n- Include all 3 requirements: MCP configured, Gecho Chrome extension installed and logged in, TikTok web logged in and tab kept open.\n- Include the OpenClaw MCP setup command and `openclaw mcp list` verification command.\n- Include the Chrome extension link.\n- Include after-setup wording that tells the user to return to OpenClaw Dashboard or Hermes and retry the request.\n- Include all related links: website, OpenClaw setup video, Hermes setup video, YouTube channel, GitHub/README, Discord, and WeCom QR code.\n- Include related Gecho Skills listed in the template.\n- Do not recommend `openclaw plugins install` in this setup-missing response.\n\n````markdown\nGecho Bridge is not ready yet.\n\nThis Skill is installed, but the official Gecho Bridge MCP tools are not available in this session yet. Installing the Skill alone does not start the TikTok search service.\n\nGecho requires all 3 items below before TikTok search or insight can run:\n\n1. Gecho Bridge MCP is configured.\n2. The Gecho Chrome extension is installed and logged in to a Gecho account.\n3. TikTok web is logged in inside Chrome, with the TikTok tab kept open.\n\nFollow these first-time setup steps:\n\n**1. Install the Gecho Chrome extension**\n\nChrome Web Store:\nhttps://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb\n\nAfter installing, open the extension and log in to your Gecho account.\n\n**2. Configure Gecho Bridge MCP**\n\nOpenClaw MCP setup:\n\n```bash\nopenclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart\n```\n\nThen verify:\n\n```bash\nopenclaw mcp list\n```\n\nFor Hermes:\n\n```bash\nhermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart\n```\n\n**3. Log in to TikTok web**\n\nOpen TikTok in Chrome, log in, and keep the logged-in TikTok tab open.\n\nAfter setup, return to OpenClaw Dashboard or Hermes and ask again, for example:\n\"Search computers on TikTok\"\n\n\n**Related links**\n\n- Website: https://gecho.ai/\n- OpenClaw setup video: https://www.youtube.com/watch?v=ggwY9hISHcQ\n- Hermes setup video: https://www.youtube.com/watch?v=zHKnuWnxt_c\n- YouTube channel: https://www.youtube.com/@Gecho-AI\n- GitHub and README: https://github.com/gecho-ai/gecho-bridge\n- Discord support: https://discord.gg/RFDVZMR6Tn\n- WeCom QR code: https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg\n\n**Related Gecho Skills**\n\n- `tiktok-video-search`: TikTok video search and metadata collection.\n- `tiktok-insight`: TikTok product, trend, competitor, and content insight jobs.\n- `tiktok-search`: complete TikTok search and insight workflow.\n````\n\n## Troubleshooting\n\n| Situation | What to do |\n|---|---|\n| MCP tools are missing | Give the setup-missing response. Do not run local shell probes. |\n| User installed only the Skill | Explain that Skill-only install is not enough and provide the OpenClaw MCP setup command. |\n| Hermes MCP tools are missing | Provide the `hermes mcp add ...` command. Do not inspect or rewrite Hermes config files. |\n| Extension not connected | Ask the user to enable/login to the Gecho Chrome extension and keep a logged-in TikTok tab open. |\n| CAPTCHA or login wall | Ask the user to resolve it manually in Chrome, then retry in a later turn. |\n| Request timeout | Report the timeout, mention stuck page/network/CAPTCHA as likely causes, and stop. |\n| Empty search results | Say the exact keyword returned no results and ask the user to choose another keyword manually. |\n| Insight still running | Report running status and recommend checking again after about 60 seconds. |\n| Failed to save results | Ask the user to provide a valid absolute directory path with write permission. |\n\n## FAQ\n\n### Why is the Chrome extension required? Can't I just use the web page?\n\nGecho needs real-time platform data from a live browser session, such as TikTok videos and other platform data in Gecho workflows. The Chrome extension connects the AI workflow to the user's logged-in Chrome session; the Skill page alone cannot collect this data.\n\n### Why do I need to log in to TikTok? Can I use it without login?\n\nTikTok limits content access for logged-out users. After you log in, the extension can access the complete data available in your browser session, such as video captions/scripts, comments, engagement data, and other signals when available.\n\nGecho does not ask for or collect your TikTok password, private account information, or publish anything on your behalf.\n\n### Need help?\n\nScan the [WeCom QR code](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg) for 1:1 support.\n\n## Output guidelines\n\nFor successful search:\n\n- Say the search completed.\n- Include total result count if available.\n- Include saved file path if available.\n- Show top 3 to 5 results only.\n- Do not paste the full raw JSON into chat.\n\nFor successful insight start:\n\n- Say the insight job started.\n- Include the `jobId`.\n- Include expected saved path if available.\n- Tell the user to check status later.\n\nFor completed insight:\n\n- Summarize key findings.\n- Include saved file path.\n- Avoid claiming conclusions not supported by returned tool data.\n\nFor failures:\n\n- Report the exact tool error or failure state.\n- Provide only the relevant fix from Troubleshooting.\n- Include the setup and support links block so the user can continue setup through docs, videos, or support.\n- Do not retry in the same turn.\n\n## Scope and limits\n\nThis Skill should:\n\n- Help users complete the official Gecho setup when prerequisites are missing.\n- Route TikTok search and insight requests to the official Gecho MCP tools.\n- Keep search, insight, and status-check flows explicit.\n- Summarize results without flooding the chat.\n\nThis Skill must never:\n\n- Pretend the Skill page alone is enough when MCP is missing.\n- Pretend `tiktok_insight` is synchronous.\n- Use unofficial TikTok scraping workflows.\n- Invent results when the tool returns no data.\n- Solve CAPTCHA, log in to TikTok, or operate the user's browser outside the official Gecho MCP workflow.\n\nFile v1.1.24:_meta.json\n\n{\n  \"ownerId\": \"kn7b732kkfejrjfkjq2h0hc9qn84nv5k\",\n  \"slug\": \"tiktok-search\",\n  \"version\": \"1.1.24\",\n  \"publishedAt\": 1780756109312\n}\n\nFile v1.1.24:skill-card.md\n\n## Description: <br>\nSearch TikTok videos and run product, trend, competitor, and content insights through Gecho Bridge MCP. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[gecho-ai](https://clawhub.ai/user/gecho-ai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nExternal users and developers use this skill to search TikTok, collect structured video metadata, and start asynchronous product, trend, or competitor insight jobs through Gecho Bridge. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: TikTok searches and insight jobs use Gecho Bridge and a logged-in Chrome/TikTok browser session. <br>\nMitigation: Install and use the skill only when you are comfortable connecting Gecho Bridge and the Gecho Chrome extension to that browser session. <br>\nRisk: Search and insight outputs may be saved locally as JSON files containing research data. <br>\nMitigation: Choose a save directory deliberately when possible and delete retained result files when they are no longer needed. <br>\nRisk: The skill depends on the official Gecho MCP tools and can fail if the MCP server, extension login, or TikTok session is unavailable. <br>\nMitigation: Configure the official Gecho Bridge MCP server, keep the extension and TikTok web logged in, and report exact tool failures instead of substituting unofficial scraping methods. <br>\n\n\n## Reference(s): <br>\n- [TikTok Search on ClawHub](https://clawhub.ai/gecho-ai/tiktok-search) <br>\n- [Gecho Publisher Profile](https://clawhub.ai/user/gecho-ai) <br>\n- [Gecho Website](https://gecho.ai/) <br>\n- [Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md) <br>\n- [Gecho Chrome Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb) <br>\n- [OpenClaw Setup Video](https://www.youtube.com/watch?v=ggwY9hISHcQ) <br>\n- [Hermes Setup Video](https://www.youtube.com/watch?v=zHKnuWnxt_c) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Text, Markdown, Shell commands, Configuration, JSON, Guidance] <br>\n**Output Format:** [Markdown guidance with bash command blocks and JSON-oriented TikTok result summaries.] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May return a jobId for asynchronous insight jobs; Gecho may save search or insight results as local JSON files.] <br>\n\n## Skill Version(s): <br>\n1.1.24 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\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. <br>\n\nArchive v1.1.23: 3 files, 6676 bytes\n\nFiles: _meta.json (133b), skill-card.md (2868b), SKILL.md (14463b)\n\nFile v1.1.23:SKILL.md\n\n---\nname: tiktok-search\ndescription: Search TikTok videos and run product, trend, competitor, and content insights through Gecho Bridge MCP. Use for TikTok keyword research, trending video discovery, winning-product analysis, and checking async insight jobs. Requires the Gecho Chrome extension, an active TikTok session, and the shared Gecho Bridge MCP server.\nmetadata:\n  openclaw:\n    os: [\"darwin\", \"linux\", \"win32\"]\n    requires:\n      bins: [\"node\", \"npx\"]\n  hermes:\n    tags: [tiktok, search, insight, gecho, mcp]\n    category: social-media\n    os: [darwin, linux, windows]\n---\n\n# TikTok Search & Insight by Gecho\n\nSearch TikTok from an AI chat, collect structured video metadata, and run async product or trend insight jobs through the official Gecho Bridge MCP workflow.\n\nThis is the default TikTok aggregate Skill for Gecho. It covers the TikTok video search and TikTok insight workflow. Single-tool Ti\n\nArchive v1.1.22: 3 files, 8245 bytes\n\nFiles: _meta.json (133b), skill-card.md (2367b), SKILL.md (20397b)\n\nArchive v1.1.21: 2 files, 6837 bytes\n\nFiles: _meta.json (133b), SKILL.md (19709b)\n\nArchive v1.1.19: 2 files, 5583 bytes\n\nFiles: _meta.json (133b), SKILL.md (14828b)","readmeExcerpt":"Skill: tiktok-search Owner: gecho-ai Summary: Search TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server. Tags: latest:1.1.38 Version history: v1.1.38 | 2026-09-23T08:58:48.634Z | user Release 1.1.38 ","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"openclaw mcp set gecho-bridge '{\"command\":\"npx\",\"args\":[\"-y\",\"@gecho-ai/gecho-bridge@latest\"]}'\nopenclaw gateway restart"},{"language":"bash","snippet":"openclaw mcp list"},{"language":"bash","snippet":"openclaw plugins install clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart"},{"language":"bash","snippet":"openclaw plugins update clawhub:@gecho-ai/gecho-bridge-bundle\nopenclaw gateway restart"},{"language":"bash","snippet":"hermes mcp add gecho-bridge --command npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart"},{"language":"bash","snippet":"hermes mcp add gecho-bridge --command /opt/homebrew/bin/npx --args=\"-y\" --args=\"@gecho-ai/gecho-bridge@latest\"\nhermes restart"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: tiktok-search\ndescription: Search TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP. Requires the Gecho Chrome extension, an active TikTok session, and the Gecho Bridge MCP server.\nmetadata:\n  openclaw:\n    os: [\"darwin\", \"linux\", \"win32\"]\n    requires:\n      bins: [\"node\", \"npx\"]\n  hermes:\n    tags: [tiktok, search, insight, gecho, mcp]\n    category: social-media\n    os: [darwin, linux, windows]\n---\n\n# TikTok Search & Insight by Gecho\n\nSearch TikTok from an AI chat, collect structured video metadata, and run async product or trend insight jobs through the official Gecho Bridge MCP workflow.\n\nThis is the default TikTok aggregate Skill for Gecho. It covers TikTok video search, creator collection, one-video detail and comment collection, and TikTok insight. Single-tool TikTok Skills may exist for distribution and search traffic, but this Skill is the recommended default for users who want the complete TikTok research workflow.\n\n## Critical prerequisite: read before use\n\nGecho Skills must be used with the Gecho Chrome extension. You must be logged in to both your Gecho account in the extension and TikTok in the Chrome web app. If either login is missing, all TikTok search and insight features may fail even though the Skill is installed.\n\n## 3-step quick start\n\n### Step 1: Install the Gecho Chrome extension\n\n1. Open the [Gecho Chrome extension download page](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb).\n2. Click `Add to Chrome`, then confirm `Add extension`.\n\n### Step 2: Log in to the Gecho extension\n\nOpen the Gecho extension in Chrome and log in to your Gecho account. Keep the extension online.\n\n### Step 3: Log in to TikTok web\n\nOpen TikTok in Chrome and log in to the TikTok web app. Keep the logged-in TikTok tab open.\n\nAfter setup is complete, return to the OpenClaw dashboard or Hermes and ask: \"Search xxx on TikTok\". For example: \"Search hamburgers on TikTok\".\n\n## Official links and setup help\n\n- Website: [gecho.ai](https://gecho.ai/)\n- GitHub: [gecho-ai/gecho-bridge](https://github.com/gecho-ai/gecho-bridge)\n- YouTube channel: [@Gecho-AI](https://www.youtube.com/@Gecho-AI)\n- Chrome extension: [Gecho Extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- OpenClaw setup video: [OpenClaw + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- Hermes setup video: [Hermes + TikTok: Direct AI Browser Control via Gecho Bridge](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n- Discord: [https://discord.gg/RFDVZMR6Tn](https://discord.gg/RFDVZMR6Tn)\n- WeCom group QR code: [qywx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/qywx.jpg)\n- 1:1 support QR code: [wx.jpg](https://github.com/gecho-ai/gecho-bridge/blob/main/wx.jpg)\n\n## What this skill does\n\n- Finds high-"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn7b732kkfejrjfkjq2h0hc9qn84nv5k\",\n  \"slug\": \"tiktok-search\",\n  \"version\": \"1.1.38\",\n  \"publishedAt\": 1790153928634\n}"},{"path":"skill-card.md","content":"## Description:\n\nSearch TikTok videos, collect creator videos, retrieve detail data and comments for one known video, and run product, trend, competitor, and content insights through Gecho Bridge MCP.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gecho-ai](https://clawhub.ai/user/gecho-ai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users, developers, and social-media researchers use this skill to search TikTok, collect video and creator metadata, inspect one known video, and run product, trend, competitor, or content insight workflows through Gecho Bridge MCP.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The workflow depends on a third-party MCP server and Chrome extension using the user's logged-in TikTok browser session.\n\nMitigation: Install and use it only when comfortable allowing that integration to read TikTok pages available in the browser session.\n\nRisk: Search, video, and insight workflows can fail when Gecho Bridge MCP, the Chrome extension, or the TikTok browser session is not ready.\n\nMitigation: Confirm Node/npx, Gecho Bridge MCP, the logged-in Gecho Chrome extension, and an active logged-in TikTok tab before use.\n\nRisk: Collected TikTok result sets may be saved locally.\n\nMitigation: Choose an appropriate save directory and handle generated files according to the user's data-handling requirements.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gecho-ai/skills/tiktok-search)\n- [Gecho website](https://gecho.ai/)\n- [Gecho Bridge README](https://github.com/gecho-ai/gecho-bridge/blob/main/README.md)\n- [Gecho Chrome extension](https://chromewebstore.google.com/detail/pjkaeenpekolahdbccjfenjcmanemlbj?utm_source=item-share-cb)\n- [OpenClaw setup video](https://www.youtube.com/watch?v=ggwY9hISHcQ)\n- [Hermes setup video](https://www.youtube.com/watch?v=zHKnuWnxt_c)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown with concise summaries, setup commands, saved file paths, and bounded result samples]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May reference JSON result files produced by Gecho MCP tools; successful insight starts return a jobId for later status checks.]\n\n## Skill Version(s):\n\n1.1.38 (source: server release metadata and artifact _meta.json)\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":null,"editorialQuality":{"score":100,"threshold":65,"status":"thin","wordCount":1451,"uniquenessScore":43,"reasons":["uniqueness-below-45"]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-10T04:10:57.267Z","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":"2026-10-10T04:10:57.267Z","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":"2026-10-10T08:47:29.021Z","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"}]}}}