@noanswer/context-compose/assets/mcps/web-search.yaml

Orchestrate complex AI interactions with Context Compose. A powerful CLI and server for building, validating, and managing context for large language models using the Model Context Protocol (MCP).

version: 1
kind: mcp
name: web-search
description: Web search capabilities for research and information gathering
prompt: |
  Conduct comprehensive research using targeted web searches.
  Optimize queries, prioritize reliable sources, and validate
  information to gather accurate and current data for any topic,
  especially for technical problem-solving and documentation.
enhanced-prompt: |
  # 🔍 Web Search Research

  ## 1. Search Strategy
  Formulate precise queries to get the most relevant results.

  **A. Query Optimization**
  Use quotes for exact phrases and site-specific searches for focus.
  ```bash
  # Search for a specific error within the official docs
  "React hooks useEffect cleanup" site:reactjs.org

  # Find best practices for a technology
  "TypeScript generic constraints" best practices
  ```

  **B. Source Prioritization**
  - **High Priority:** Official documentation, reputable tech blogs.
  - **Medium Priority:** GitHub issues, Stack Overflow.
  - **Low Priority:** Unverified forums, outdated articles.

  ## 2. Information Validation
  Always critically evaluate the information you find.
  - **Check Recency:** Is the information up-to-date? Look for publication dates.
  - **Verify Authority:** Is the source credible? (e.g., official docs vs. personal blog).
  - **Cross-Reference:** Can you confirm the information from multiple trusted sources?
  - **Test Code:** If you find code snippets, test them to ensure they work as expected.

  ## 3. Common Search Categories
  - **Troubleshooting:** `"Error message" "technology" fix`
  - **Best Practices:** `"[technology] best practices 2024"`
  - **Tutorials:** `"[technology] getting started guide"`

  **🎯 Result:** Comprehensive and validated research with actionable insights from reliable sources.