tool_search_web
Search the web using DuckDuckGo to find relevant information, articles, and resources for development tasks. Returns structured results with titles, URLs, and snippets for efficient research.
Instructions
Search the web using DuckDuckGo.
Args: query: Search query string. limit: Maximum results (1-20, default 5).
Returns: List of results with title, url, snippet.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| limit | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Implementation Reference
- src/devlens/server.py:40-51 (registration)The tool `tool_search_web` is defined and registered here using the `@mcp.tool()` decorator. It delegates to the `search_web` helper function.
@mcp.tool() async def tool_search_web(query: str, limit: int = 5) -> list[dict]: """Search the web using DuckDuckGo. Args: query: Search query string. limit: Maximum results (1-20, default 5). Returns: List of results with title, url, snippet. """ return await search_web(query, limit) - src/devlens/tools/search.py:20-74 (handler)The implementation of the web search logic, using `DDGAdapter` to perform the actual search.
async def search_web( query: str, limit: int = 5, *, region: str | None = None, safe_search: bool = True, ) -> list[dict]: """Search the web using DuckDuckGo. Args: query: Search query string. Supports operators like: - site:domain.com to search specific domain - "exact phrase" for exact matches - -word to exclude terms limit: Maximum number of results (1-20). region: Optional region code (e.g., 'us-en', 'uk-en'). safe_search: Enable safe search filtering. Returns: List of search results with title, url, and snippet. Raises: SearchError: If search fails or query is invalid. Example: >>> results = await search_web("Python asyncio tutorial", limit=5) >>> results = await search_web('site:docs.python.org async', limit=3) """ if not query or not query.strip(): raise SearchError(query, "Query cannot be empty") # Normalize query normalized = _normalize_query(query) # Clamp limit limit = min(max(limit, 1), 20) try: results = await _adapter.search( normalized, limit=limit, region=region, safe_search=safe_search, ) # Filter out low-quality results filtered = [r for r in results if r.title and r.url and len(r.snippet) > 20] return [r.model_dump() for r in filtered] except SearchError: raise except Exception as e: raise SearchError(query, f"Unexpected error: {e}") from e