Web Search Intake Governance
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @user_15292d5a/yjkj-web-search-rules.
About this skill
Problem
Web search results are not automatically safe knowledge-base material. Saving everything mixes in low-trust pages, duplicate URLs, promotional bodies, and sensitive domains, making cleanup harder. web-search-rules treats the search-to-intake path as an auditable workflow instead of letting the model treat webpage text as trusted instructions.
How It Works
The skill loads rules from ~/.skill-config/web-search-rules/config.json, then normalizes URLs, deduplicates results, and classifies each source. Trust levels include trusted, allowed, review, and blocked, with legacy whitelist and blacklist terms mapped for compatibility. Rule types support exact_url, domain, path_prefix, keyword, topic, and source_type; conflicts are resolved with blocked first and user confirmation. Intake actions are separate: allow_stage, allow_archive, allow_cloud_upload, needs_review, and blocked. Cloud uploads, deletion, and migration require dry-run reports and second confirmation.
Boundaries
This is suitable for governing web material before adding it to Obsidian, Feishu, Tencent Docs, DingTalk Docs, or custom platforms. It is not for automatic login, protected scraping, or unconfirmed bulk uploads. Webpage content remains untrusted input; config must not store passwords, cookies, tokens, or API keys. NotebookLM, browser automation, deletion, and migration raise the risk level.
Use Cases
- When researching competitors, stage web results locally and classify source trust before writing to the knowledge base.
- Before using Feishu Wiki, run blocked rules on candidate sources and create a review list to prevent low-quality pages from being archived.
- When migrating legacy web-search-rules config, inspect old paths read-only and produce a dry-run report before creating the new config.
- When curating an Obsidian vault, stage search hits as Markdown and confirm archiving under topic rules.
Best For
- Knowledge managers who want web material staged and trust-classified before archiving into a personal knowledge base.
- Engineering teams who need reviewed external sources before adding them to Feishu or Tencent Docs.
- Compliance-minded architects who require dry-run reports and second confirmation for deletion, migration, and cloud upload.
- Obsidian researchers who want to filter webpage summaries by source trust and keep audit records.
Related Skills
Reads conversation-trace files to generate an animal- or mythology-based soul mirror card and Johari Window insights.
A Deling knowledge-base research workflow that clarifies intent, runs broad and vertical searches, supplements with web sources, validates diversity, and traces key claims.
A SiYuan knowledge-base management skill for double-link parent indexes, MOCs, numbered documents, tags, repo sync, and WeChat import workflows.
A structured workflow for academic literature reviews, covering multi-database search, screening, thematic synthesis, citation validation, and PDF output.