Our review
Performs web searches, documentation lookups, and technology evaluations to provide current and synthesized information.
Strengths
- Access to official and recent sources
- Structured synthesis with recommendations
- Automatic saving of important research
Limitations
- Requires web access via Claude Code
- Quality depends on available sources
- May need clarification on the research question
Use this skill when you need information from outside the codebase, such as API documentation, technology comparisons, or best practices.
Avoid using it for exploring existing project code, implementing features, or obtaining information already present in the codebase.
Security analysis
SafeThe skill only uses WebSearch, WebFetch, Read, Write to look up information and save markdown files. No destructive, exfiltration, or obfuscated commands are present.
No concerns found
Examples
Should we use Redis or Memcached for caching? Compare their features, performance, and trade-offs for our project.How does Stripe webhook verification work? Show me the steps and provide a code example in Python.name: researching-topics description: > Web search, documentation lookup, and technology evaluation. Use when user says "research", "look up", "find docs", "what is", "compare options", "evaluate", or needs information from outside the codebase. Do NOT use for: codebase exploration (use exploring-codebase), implementing features (use implementing-features). compatibility: "Requires Claude Code with web access." allowed-tools: [WebSearch, WebFetch, Read, Write] metadata: author: agentic-framework version: "${VERSION}"
Researching Topics
Web search, documentation lookup, and technology evaluation.
Instructions
Step 1: Clarify the Research Question
Understand what information the user needs:
- Specific API documentation?
- Technology comparison?
- Best practices for a technique?
- Bug/error resolution?
Step 2: Search and Gather
Use web search for current information:
- Official documentation first
- Stack Overflow for common issues
- GitHub issues for library-specific problems
- Blog posts for best practices (prefer recent sources)
Step 3: Synthesize Findings
Present findings organized by relevance:
- Direct answer to the question
- Supporting evidence and sources
- Trade-offs or alternatives if applicable
- Recommended approach with justification
Step 4: Save if Valuable
If the research informs a decision, save it:
docs/research/YYYY-MM-DD-topic.md
Examples
Example 1: Technology evaluation User says: "Should we use Redis or Memcached for caching?" Steps taken:
- Search for current comparisons and benchmarks
- Check project's existing infrastructure (STACK.md)
- Compare: persistence, data structures, clustering, ease of setup Result: Recommendation with trade-offs table and links to sources.
Example 2: API documentation lookup User says: "How does the Stripe webhook verification work?" Steps taken:
- Fetch Stripe's official webhook docs
- Extract key steps: signature verification, event handling
- Provide code example matching project's language Result: Step-by-step guide with code snippet and security notes.
Troubleshooting
Outdated information found Cause: Search results may include old articles. Solution: Check publication dates. Prefer official docs. Cross-reference multiple sources.
API Documentation Generator
Documentation
Automatically generates OpenAPI/Swagger API documentation.
Technical Writer
Documentation
Writes clear technical documentation following top style guides.
Pivot Decision Framework
Documentation
Documents a strategic pivot or persevere decision with evidence, analysis, and rationale. Use when evaluating whether to change direction on a product, feature, or strategy based on market feedback.