championswimmer/tech-doc-skills/web-research
Claude Code and agent skill by championswimmer/tech-doc-skills
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Use this skill for current, source-backed technical research for docs, RFCs, ADRs, release notes, API usage, benchmarks, comparisons, or ecosystem questions. Best when Exa, Parallel Search, Perplexity, and Context7 MCP tools are available.
Install
npx vskill@latest install championswimmer/tech-doc-skills/web-researchSKILL.md
Use this skill for current technical research with citations.
Tool map
Prefer these MCP tools by name when available:
exa:web_search_exa— first-pass semantic discovery; best for official docs, product pages, release notes, and likely canonical URLsexa:web_fetch_exa— read chosen URLs as markdownparallel-search:web_search— run 3-4 distinct query angles in parallel for breadthparallel-search:web_fetch— fetch full text for URLs found via parallel searchperplexity:perplexity_search— ranked URL discovery when Exa is weak or unavailableperplexity:perplexity_ask— quick synthesis or cross-check with citations after discoverycontext7:resolve-library-id->context7:query-docs— library/framework/API docs and code examples
For Claude Code permission syntax, the same tools are usually named mcp__server__tool, for example mcp__exa__web_search_exa.
Default loop
- Restate the research goal in one sentence.
- Write 3-5 search angles: official docs, source repo, release notes, comparison, caveats.
- Start with
exa:web_search_exa. - Use
parallel-search:web_searchwhen one phrasing may miss results. - Read real pages with
exa:web_fetch_exaorparallel-search:web_fetch; do not rely only on snippets. - Use
perplexity:perplexity_askonly after reading enough sources to synthesize or cross-check. - If the question is library-specific, call
context7:resolve-library-idthencontext7:query-docs. - Cite every non-obvious claim and mark uncertainty.
Routing rules
- library or API usage -> Context7 first, then web sources for release notes or ecosystem context
- official docs / changelogs / vendor pages -> Exa first
- broad comparisons or ecosystem scans -> Parallel Search first, then fetch
- quick factual cross-check -> Perplexity ask
- code-heavy topic -> add source repo and API docs; prefer implementation over blogs
Minimum evidence standard
- consequential claim -> 1 official source + 1 corroborating source or implementation reference
- freshness-sensitive claim -> include version or date
- disagreement -> show both sides; do not average them away
References
Load only as needed:
references/query-patterns.mdreferences/tool-failover.mdreferences/source-quality.mdreferences/rfc-research-output-template.md
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/web-research/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/championswimmer/tech-doc-skills/web-research)