Web Collection
云端优先的网页采集、上手与排障
Rank
62
Safety
84
Downloads
2.4k
Updated
Oct 9, 2026
Version
1.2.9
Source
CLAWHUB
About
What it does, and when to use it.
Capability contract not published. No trust telemetry is available yet. 2.4K downloads reported by the source. Last updated 10/9/2026.
Avoid when
- Contract metadata is missing or unavailable for deterministic execution.
Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing
Public facts
Every fact links back to the source it came from.
- Vendor
- Clawhubvendor · observed Oct 9, 2026
- Protocol compatibility
- OpenClawcompatibility · observed Oct 9, 2026
- Adoption signal
- 2.4K downloadsadoption · observed Oct 9, 2026
- Latest release
- 1.2.9release · observed Jun 28, 2026
- Handshake status
- UNKNOWNsecurity
Install and run
Setup complexity: low.
clawhub skill install s17973ckq18hmgn6fyfhwykv1583h5nd:web-collection- Install using `clawhub skill install s17973ckq18hmgn6fyfhwykv1583h5nd:web-collection` 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/yiming1001/web-collection before using production credentials.
Contract: missing
curl -s "https://www.xpersona.co/api/v1/agents/clawhub-yiming1001-web-collection/snapshot"
Documentation
CLAWHUB
149,215 characters of source documentation, loaded on request.
Extracted files
5 files captured from the source.
SKILL.md
--- name: web-collection description: 通过云端连接器优先执行浏览器插件数据采集,也可回退到本地连接器;适用于抖音、TikTok、小红书、Amazon、Bilibili 的采集任务,以及 web-collection 首次上手、配置、付费使用说明和 QA 排障。 --- # Web Collection Use this skill for browser-extension collection tasks on: - Douyin - TikTok - Xiaohongshu - Amazon - Bilibili ## Online Documentation Routing The online documents are the source of truth for user-facing guidance. Do not duplicate their long-form onboarding, paid-access, UI operation, or QA content in this skill. When a user asks for guidance, read the relevant online document if tools can access it; if not, give the user the link and say the online document should be used as the current guide. - Quick start / first-time use / UI operation guide: - https://vcn5grhrq8y0.feishu.cn/wiki/MoXrwwUN7iiUFkk8eWycs9JqntA?from=from_copylink - Use when the user asks how to use the plugin, how to start, how to configure it for the first time, where to click, how paid access works, or needs step-by-step human instructions. - Demo scenarios / example collection scenes: - https://vcn5grhrq8y0.feishu.cn/wiki/GO11wlXkriSwNakXrt2ck0GanEe - Use when introducing what this skill can do, when the user asks for examples, or after a successful run to point the user to more collection scenarios. - Browser extension and connector installation guide: - https://vcn5grhrq8y0.feishu.cn/wiki/R6f2w6o7ci1db1kYLK4cgJIYnWh - Use before the user's first collection run, or when the user needs to install, download, update, or reconnect the browser extension or connector. - The browser extension and connector must be downloaded from this document. Do not tell users to search or install the extension from the Chrome Web Store / Google Store. - Bitable configuration, connector verification, and authorization guide: - https://vcn5grhrq8y0.feishu.cn/wiki/EAtJw2irFiDvMpkZXb4cBjYonNg - Use after installation is complete and before first collection. It covers plugin-side bitable configuration, connector status verification, and manual fallback details. - Bitable template direct link: - https://vcn5grhrq8y0.feishu.cn/base/UKQsbVHpMac293s0cnFc1hq1nDd?table=tblTXM4lclXM6Jzr&view=vew8OdcKHw - When guiding the user to copy the bitable template, send this direct link in the response so the user does not need to open the guide first to find it. - QA / troubleshooting guide: - https://vcn5grhrq8y0.feishu.cn/wiki/F83hw2w6Xi7EOFkIScrccrQYnnd?from=from_copylink - Use when the user asks about errors, failed export, missing data, connector/cloud/local issues, access problems, supported scenarios, or other common questions. Decision rule: - If the user asks what this skill is, what it can do, or generally says "how do I use this skill" without explicitly asking to configure now, give only the lightweight intro response. Do not show the full operation flow, setup checklist, or screenshots yet. End by asking whether they want to continue with first-time setup/configuration. - If the user confirms they want t
assets/README.md
# Web Collection Visual Assets Place optional onboarding screenshots or GIFs in this folder. The online Feishu documents remain the source of truth; these files are for quick in-chat visual guidance. Agents should resolve these images relative to the skill package, from the bundled `assets/` directory. Use the image/attachment mechanism supported by the current Agent host. If Markdown image rendering for packaged relative assets is supported, use: ```markdown  ``` If the host requires explicit attachments, attach the matching file from `assets/`. Do not hardcode machine-specific absolute paths in user-facing guidance. Do not reference these screenshots as plain text only when the user is asking for setup guidance. For configuration guidance, images should be placed next to their matching text step, not collected at the bottom of the response. Bitable binding uses one image per substep. The connector image is the only exception: it is a single overview image for the whole third major step and appears once at the end of that major step. Recommended filenames: - `install-extension-connector.gif` or `install-extension-connector.png` - Browser extension and connector installation. - `bitable-step-00-personal-center.png` - Open the browser extension, enter `个人中心`, and log in if needed. - `bitable-step-01-template-copy.png` - Open the direct bitable template link, create a copy, and choose the proper copy settings. - `bitable-step-02-auth-code.png` - Open `多维表格插件` -> `自定义插件`, enable the authorization code, and copy it. - `bitable-step-03-config-save.png` - Fill in the bitable link and personal authorization code, test the connection, and save. - `connector-step-01-status-token.png` - Overview image for the whole third major step: open `http://127.0.0.1:19820`, authorize login, verify all green statuses, and copy device ID plus Token. Use this screenshot exactly once at the end of connector guidance; do not repeat it under separate status verification and credential copying substeps. - `view-results.gif` or `view-results.png` - Where to view exported bitable or CSV results. Downloaded from the bitable configuration / connector verification guide: - Guide URL: `https://vcn5grhrq8y0.feishu.cn/wiki/EAtJw2irFiDvMpkZXb4cBjYonNg` - Direct bitable template URL: `https://vcn5grhrq8y0.feishu.cn/base/UKQsbVHpMac293s0cnFc1hq1nDd?table=tblTXM4lclXM6Jzr&view=vew8OdcKHw` - `bitable-step-00-personal-center.png` - `bitable-step-01-template-copy.png` - `bitable-step-02-auth-code.png` - `bitable-step-03-config-save.png` - `connector-step-01-status-token.png` Keep media small enough for fast Skill installation. Prefer short GIFs or compressed PNG/JPG screenshots. Put large videos in the Feishu knowledge base and link to them from `SKILL.md`.
README.md
# web-collection `web-collection` is a browser-extension data collection skill for Douyin, TikTok, Xiaohongshu, Amazon, and Bilibili. It prioritizes the Agent-led connector flow, with local connector execution as a troubleshooting fallback. The skill covers first-time onboarding, connector and bitable setup, website login confirmation for connector authorization, closed-loop execution, export handling, and troubleshooting. ## Core Files - `SKILL.md` - `scripts/run.sh` - `scripts/preflight_check.sh` - `scripts/ensure_connector_auth.sh` - `scripts/cloud_dispatch_loop.sh` - `scripts/collect_and_export_loop.sh` - `scripts/export_preference.sh` - `scripts/reexport_task.sh` - `references/learning-guide.md` - `assets/` ## Install Install only this skill directory: ```bash python3 <skill-installer-path>/install-skill-from-github.py \ --url https://github.com/yiming1001/skills-yiming/tree/main/web-collection ``` The installed directory should preserve this structure: ```text web-collection/ SKILL.md scripts/ references/ assets/ ``` Restart your coding agent or client after installation so the skill is loaded. ## Execution Entry Use the wrapper script: ```bash bash scripts/run.sh --platform douyin --keyword "AI" ``` The wrapper reads stored preferences, resolves connector authorization from existing state or a website login confirmation link, runs preflight checks, chooses exactly one internal execution mode, and dispatches through either the cloud connector loop or the local connector loop.
_meta.json
{
"ownerId": "kn7e8rp9zay3je2bycn9zrmztx82hy57",
"slug": "web-collection",
"version": "1.2.9",
"publishedAt": 1782616528090
}references/learning-guide.md
# Web Collection Learning Guide This file is an offline routing and recovery summary for the agent. The online documents are the source of truth for user-facing instructions and QA: - Quick start / first-time use / UI operation guide: https://vcn5grhrq8y0.feishu.cn/wiki/MoXrwwUN7iiUFkk8eWycs9JqntA?from=from_copylink - QA / troubleshooting guide: https://vcn5grhrq8y0.feishu.cn/wiki/F83hw2w6Xi7EOFkIScrccrQYnnd?from=from_copylink When a user asks how to use the paid plugin, how to configure it for the first time, where to click, or how access works, read or point them to the quick-start guide. When a user asks about errors, export failures, missing data, connector status, local/cloud behavior, or common questions, read or point them to the QA guide. Use the rest of this file only to keep execution safe when online docs are unavailable or when running the bundled scripts. Use one unified rule set: 1. Never ask for configuration that is already present in environment variables. 2. Do not ask the user to manually copy connector credentials during normal Agent-led use. 3. Connector authorization is resolved by `scripts/ensure_connector_auth.sh` from explicit values, App state, connector state, or a generated website login confirmation link. 4. Local and cloud are internal execution modes, not user-facing setup choices. 5. Local mode may only call `scripts/collect_and_export_loop.sh`. 6. Cloud mode may only call `scripts/cloud_dispatch_loop.sh`. 7. Advanced filters, including time-based filters, must be passed through inside `payload.filters`. 8. Do not invent one universal filter schema in this skill; different platforms and methods may require different filter keys or value formats. 9. If the exact filter shape is unclear, check `GET /api/filters` first before guessing. 10. Deduplication fields are resolved by the connector/plugin, never by this skill. 11. Personal bitable export must remain `exportMode=personal`; deduplication is what makes the new plugin choose its smart personal export internally. ## Asking Rules Ask only for user-facing defaults that are still missing after checking: - `defaultExportMode` - `defaultMaxItems` - `defaultFetchDetail` - `defaultDetailSpeed` - `defaultDeduplicationEnabled` - `defaultDeduplicationStrategy` Do not ask for: - deduplication field or any field-selection input - `WEB_COLLECTION_CONNECTION_MODE` - `WEB_COLLECTION_BRIDGE_URL` - `WEB_COLLECTION_CLOUD_BASE_URL` - `WEB_COLLECTION_CLOUD_DEVICE_ID` - `WEB_COLLECTION_CLOUD_TOKEN` - `WEB_COLLECTION_BRIDGE_CMD` ## Shared Flow 1. Run the collection entry point, which calls `scripts/ensure_connector_auth.sh` 2. Persist any user-provided defaults if this turn supplies them 3. If authorization is required, give the generated website login confirmation link and ask the user to click it 4. Build the payload 5. If the user asked for time filtering or other advanced filtering, put it inside `filters` and preserve the platform/method-native shape 6. Dispatch through
AionUi
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!
activepieces
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
cherry-studio
AI productivity studio with smart chat, autonomous agents, and 300+ assistants.
CopilotKit
The Frontend for Agents & Generative UI. React + Angular
Machine-readable data
The same record, as JSON, for agents and crawlers.
{
"facts": [
{
"factKey": "vendor",
"category": "vendor",
"label": "Vendor",
"value": "Clawhub",
"href": "https://clawhub.ai/yiming1001/skills/web-collection",
"sourceUrl": "https://clawhub.ai/yiming1001/skills/web-collection",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-09T14:59:13.241Z",
"isPublic": true
},
{
"factKey": "protocols",
"category": "compatibility",
"label": "Protocol compatibility",
"value": "OpenClaw",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-yiming1001-web-collection/contract",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-yiming1001-web-collection/contract",
"sourceType": "contract",
"confidence": "medium",
"observedAt": "2026-10-09T14:59:13.241Z",
"isPublic": true
},
{
"factKey": "traction",
"category": "adoption",
"label": "Adoption signal",
"value": "2.4K downloads",
"href": "https://clawhub.ai/yiming1001/web-collection",
"sourceUrl": "https://clawhub.ai/yiming1001/web-collection",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-09T14:59:13.241Z",
"isPublic": true
},
{
"factKey": "latest_release",
"category": "release",
"label": "Latest release",
"value": "1.2.9",
"href": "https://clawhub.ai/yiming1001/web-collection",
"sourceUrl": "https://clawhub.ai/yiming1001/web-collection",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-06-28T03:15:28.090Z",
"isPublic": true
},
{
"factKey": "handshake_status",
"category": "security",
"label": "Handshake status",
"value": "UNKNOWN",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-yiming1001-web-collection/trust",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-yiming1001-web-collection/trust",
"sourceType": "trust",
"confidence": "medium",
"observedAt": null,
"isPublic": true
}
],
"events": [
{
"eventType": "release",
"title": "Release 1.2.9",
"description": "Add onboarding screenshots, Agent-led connector authorization, direct bitable template guidance, and updated troubleshooting/re-export flow.",
"href": "https://clawhub.ai/yiming1001/web-collection",
"sourceUrl": "https://clawhub.ai/yiming1001/web-collection",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-06-28T03:15:28.090Z",
"isPublic": true
}
]
}Record generated Oct 10, 2026.
