RTFD (Read The F*****g Docs)

Instructions

Implementation Reference

• src/RTFD/providers/gcp.py:610-641 (handler)

  The main handler function for the 'search_gcp_services' tool. It invokes internal search logic and returns serialized results.
  async def search_gcp_services(query: str, limit: int = 5) -> CallToolResult: """ Search for GCP (Google Cloud Platform) services and documentation. USE THIS WHEN: You need to find Google Cloud services, APIs, or documentation for a specific GCP topic. BEST FOR: Discovering which GCP services exist for a use case or finding service documentation. Returns multiple matching services with names, descriptions, API endpoints, and docs URLs. Searches: 1. Local service mapping (exact and partial matches) 2. cloud.google.com website (fallback for specific queries) 3. googleapis GitHub repository (API definitions) After finding a service, use: - fetch_gcp_service_docs() to get full documentation content - The docs_url with WebFetch for external documentation Note: GitHub API search (fallback) is limited to 60 requests/hour without GITHUB_TOKEN. Args: query: Service name or keywords (e.g., "storage", "vertex ai", "gke audit", "bigquery") limit: Maximum number of results (default 5) Returns: JSON with list of matching services including name, description, API endpoint, docs URL Example: search_gcp_services("vertex ai") → Finds Vertex AI service with docs links """ result = await self._search_services(query, limit=limit) return serialize_response_with_meta(result)
• src/RTFD/providers/gcp.py:607-677 (registration)

  The get_tools method registers 'search_gcp_services' (and conditionally 'fetch_gcp_service_docs') in the tools dictionary for MCP.
  def get_tools(self) -> dict[str, Callable]: """Return MCP tool functions.""" async def search_gcp_services(query: str, limit: int = 5) -> CallToolResult: """ Search for GCP (Google Cloud Platform) services and documentation. USE THIS WHEN: You need to find Google Cloud services, APIs, or documentation for a specific GCP topic. BEST FOR: Discovering which GCP services exist for a use case or finding service documentation. Returns multiple matching services with names, descriptions, API endpoints, and docs URLs. Searches: 1. Local service mapping (exact and partial matches) 2. cloud.google.com website (fallback for specific queries) 3. googleapis GitHub repository (API definitions) After finding a service, use: - fetch_gcp_service_docs() to get full documentation content - The docs_url with WebFetch for external documentation Note: GitHub API search (fallback) is limited to 60 requests/hour without GITHUB_TOKEN. Args: query: Service name or keywords (e.g., "storage", "vertex ai", "gke audit", "bigquery") limit: Maximum number of results (default 5) Returns: JSON with list of matching services including name, description, API endpoint, docs URL Example: search_gcp_services("vertex ai") → Finds Vertex AI service with docs links """ result = await self._search_services(query, limit=limit) return serialize_response_with_meta(result) async def fetch_gcp_service_docs(service: str, max_bytes: int = 20480) -> CallToolResult: """ Fetch actual documentation content for a GCP (Google Cloud Platform) service. USE THIS WHEN: You need detailed documentation, guides, tutorials, or API reference for a GCP service. BEST FOR: Getting complete documentation with setup instructions, usage examples, and API details. Better than using curl or WebFetch because it: - Automatically extracts relevant content from cloud.google.com - Converts HTML to clean Markdown format - Prioritizes important sections (Overview, Quickstart, API Reference) - Removes navigation, ads, and other non-content elements - Handles multi-word service names (e.g., "gke audit policy") Works with: - Exact service names (e.g., "Cloud Storage", "Compute Engine") - Common abbreviations (e.g., "GCS", "GKE", "BigQuery") - Multi-word queries (e.g., "gke audit policy configuration") Args: service: Service name or topic (e.g., "Cloud Storage", "vertex ai", "gke audit") max_bytes: Maximum content size, default 20KB (increase for comprehensive docs) Returns: JSON with documentation content, size, source URL, truncation status Example: fetch_gcp_service_docs("vertex ai") → Returns formatted documentation from cloud.google.com """ result = await self._fetch_service_docs(service, max_bytes) return serialize_response_with_meta(result) tools = {"search_gcp_services": search_gcp_services} if is_fetch_enabled(): tools["fetch_gcp_service_docs"] = fetch_gcp_service_docs return tools
• src/RTFD/providers/gcp.py:149-162 (registration)

  The get_metadata method lists 'search_gcp_services' in tool_names for provider registration.
  def get_metadata(self) -> ProviderMetadata: tool_names = ["search_gcp_services"] if is_fetch_enabled(): tool_names.append("fetch_gcp_service_docs") return ProviderMetadata( name="gcp", description="Google Cloud Platform API and service documentation", expose_as_tool=True, tool_names=tool_names, supports_library_search=True, required_env_vars=[], optional_env_vars=["GITHUB_TOKEN", "GITHUB_AUTH"], )
• src/RTFD/providers/gcp.py:226-334 (helper)

  Core helper method implementing the search logic: local mapping, normalization, web search fallback, and GitHub API search.
  async def _search_services(self, query: str, limit: int = 5) -> list[dict[str, Any]]: """ Search for GCP services using GitHub API and local mapping. Args: query: Search query (e.g., "storage", "compute", "bigquery", "gke audit") limit: Maximum number of results to return Returns: List of matching services with metadata """ results: list[dict[str, Any]] = [] # First, check if query matches any known service normalized = self._normalize_service_name(query) if normalized and normalized in GCP_SERVICE_DOCS: service_info = GCP_SERVICE_DOCS[normalized] results.append( { "name": service_info["name"], "description": service_info["description"], "api": service_info["api"], "docs_url": service_info["url"], "source": "gcp_mapping", } ) # Search our local mapping for partial matches query_lower = query.lower() # If query has multiple words (e.g., "gke audit"), try first word as service query_words = query_lower.split() if len(query_words) > 1 and not results: # Try to find service from first word first_word_normalized = self._normalize_service_name(query_words[0]) if first_word_normalized and first_word_normalized in GCP_SERVICE_DOCS: service_info = GCP_SERVICE_DOCS[first_word_normalized] # Add note about topic in description topic = " ".join(query_words[1:]) results.append( { "name": service_info["name"], "description": f"{service_info['description']} (searching for: {topic})", "api": service_info["api"], "docs_url": service_info["url"], "source": "gcp_mapping_contextual", } ) # Search for partial matches in service names and descriptions for key, service_info in GCP_SERVICE_DOCS.items(): # Skip if already added if normalized == key: continue if len(query_words) > 1 and self._normalize_service_name(query_words[0]) == key: continue # Check if ANY word in query matches service name, key, or description query_matches = False for word in query_words: if len(word) > 2: # Skip very short words if ( word in service_info["name"].lower() or word in key or word in service_info["description"].lower() ): query_matches = True break if query_matches: results.append( { "name": service_info["name"], "description": service_info["description"], "api": service_info["api"], "docs_url": service_info["url"], "source": "gcp_mapping", } ) if len(results) >= limit: break # If we have results from local mapping (either exact or partial matches), # prioritize those over external searches if results: return results[:limit] # Only search cloud.google.com if we don't have local matches # This is for very specific queries that aren't in our mapping try: cloud_results = await self._search_cloud_google_com(query, limit) if cloud_results: results.extend(cloud_results) except Exception: pass if len(results) >= limit: return results[:limit] # Finally, try GitHub API search for googleapis repository try: github_results = await self._search_github_googleapis(query, limit) results.extend(github_results) except Exception: pass return results[:limit]
• src/RTFD/providers/gcp.py:610-638 (schema)

  Type hints and docstring define the input schema (query: str, limit: int=5) and output format for the tool.
  async def search_gcp_services(query: str, limit: int = 5) -> CallToolResult: """ Search for GCP (Google Cloud Platform) services and documentation. USE THIS WHEN: You need to find Google Cloud services, APIs, or documentation for a specific GCP topic. BEST FOR: Discovering which GCP services exist for a use case or finding service documentation. Returns multiple matching services with names, descriptions, API endpoints, and docs URLs. Searches: 1. Local service mapping (exact and partial matches) 2. cloud.google.com website (fallback for specific queries) 3. googleapis GitHub repository (API definitions) After finding a service, use: - fetch_gcp_service_docs() to get full documentation content - The docs_url with WebFetch for external documentation Note: GitHub API search (fallback) is limited to 60 requests/hour without GITHUB_TOKEN. Args: query: Service name or keywords (e.g., "storage", "vertex ai", "gke audit", "bigquery") limit: Maximum number of results (default 5) Returns: JSON with list of matching services including name, description, API endpoint, docs URL Example: search_gcp_services("vertex ai") → Finds Vertex AI service with docs links """