Mcp research
Agent skills
npx -y skills add ahgraber/skills --skill mcp-researchAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use when tasks require current, source-backed technical information from MCP tools. Apply for library/API questions, dependency version checks, third-party integration work, framework- or SDK-specific debugging, and any case where stale model knowledge could cause incorrect guidance.
SKILL.md
2.7 KB, as published. Nobody here has run it
MCP Docs and Research (Context7, Exa, Jina)
Invocation Notice
- Inform the user when this skill is being invoked by name:
mcp-research.
Overview
Use MCP-provided tools to retrieve current, verifiable information instead of relying on memory for fast-changing libraries, APIs, and ecosystem guidance.
When to Use
- Working with any external library or framework (for example, FastAPI, SQLAlchemy, pandas, boto3, or requests).
- Installing or upgrading dependencies and verifying current versions or migration guidance.
- Implementing features tied to third-party SDKs or APIs.
- Debugging behavior that may be version-specific.
- Looking up current best practices, changelogs, or breaking changes.
Tool Selection
- Use
mcp__context7__resolve-library-idandmcp__context7__query-docsfor official library documentation and API usage. - Use
mcp__exa__get_code_context_exafor code-centric examples across docs, GitHub, and Stack Overflow. - Use
mcp__exa__web_search_exafor broader current web context (announcements, release notes, ecosystem updates). - Use
mcp__jina__search_webto discover relevant pages, thenmcp__jina__read_urlfor clean page extraction. - Use
mcp__jina__search_arxivandmcp__jina__extract_pdfonly when the task needs paper-level or PDF-structured research.
Default Workflow
- Classify the request: official API docs, implementation examples, or broad web research.
- Start with the narrowest reliable source:
- Official docs first (
Context7) for API correctness. - Add
Exa/Jinaonly when you need cross-source confirmation or broader context.
- Official docs first (
- For Context7 docs, always resolve the library id before querying docs unless the exact
/org/projectid is already provided. - Keep queries specific (library + feature + version/error) to reduce noisy results.
- Synthesize findings and clearly separate sourced facts from inferences.
Quality Rules
- Prefer primary/official documentation for API signatures and behavior.
- For dependency/version decisions, verify with current documentation before recommending versions.
- Avoid unsupported claims; cite concrete tool findings.
- If sources conflict, report the conflict and recommend the safest path (pin version, test in isolation, or check release notes).
- If coverage is weak, state limits explicitly and proceed with best available evidence.