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

Instructions

Implementation Reference

• src/RTFD/providers/gcp.py:642-672 (handler)

  The main handler function implementing the fetch_gcp_service_docs MCP tool. It invokes the private helper to fetch, parse, and format GCP service documentation, then serializes the result as CallToolResult.
  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)
• src/RTFD/providers/gcp.py:673-677 (registration)

  Registration of the fetch_gcp_service_docs tool function in the dictionary returned by get_tools(), conditionally based on fetch being enabled.
  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:150-152 (registration)

  The tool name 'fetch_gcp_service_docs' is conditionally added to the list of exposed tool names in the ProviderMetadata returned by get_metadata().
  tool_names = ["search_gcp_services"] if is_fetch_enabled(): tool_names.append("fetch_gcp_service_docs")
• src/RTFD/providers/gcp.py:478-606 (helper)

  The private helper method _fetch_service_docs containing the core logic for normalizing service names, fetching HTML from GCP docs URLs, parsing with BeautifulSoup, converting to Markdown, prioritizing sections, truncating content, and handling errors.
  async def _fetch_service_docs(self, service: str, max_bytes: int = 20480) -> dict[str, Any]: """ Fetch documentation for a specific GCP service. Args: service: Service name (e.g., "storage", "compute", "Cloud Storage") max_bytes: Maximum content size in bytes Returns: Dict with content, size, source info """ try: # Normalize service name normalized = self._normalize_service_name(service) # If not found in mapping, try to construct URL if normalized and normalized in GCP_SERVICE_DOCS: service_info = GCP_SERVICE_DOCS[normalized] docs_url = service_info["url"] service_name = service_info["name"] else: # Try to search for the service search_results = await self._search_services(service, limit=1) if search_results: # Use the best match best_match = search_results[0] docs_url = best_match["docs_url"] service_name = best_match["name"] else: # Try to construct URL from service name as a last resort service_slug = service.lower().replace(" ", "-").replace("_", "-") docs_url = f"https://cloud.google.com/{service_slug}/docs" service_name = service # Fetch and parse HTML documentation headers = {"User-Agent": USER_AGENT} async with await self._http_client() as client: resp = await client.get(docs_url, headers=headers) resp.raise_for_status() soup = BeautifulSoup(resp.text, "html.parser") # Extract main documentation content # Try to find main content area # GCP docs typically use <main> or specific div classes main_content = soup.find("main") if not main_content: main_content = soup.find("div", class_=["devsite-article-body"]) if not main_content: main_content = soup.find("article") if main_content: # Remove navigation, sidebar, and other non-content elements for unwanted in main_content.find_all(["nav", "aside", "footer", "header"]): unwanted.decompose() # Remove script and style tags for script in main_content.find_all(["script", "style"]): script.decompose() # Convert to markdown html_content = str(main_content) markdown_content = html_to_markdown(html_content, docs_url) # Extract and prioritize sections sections = extract_sections(markdown_content) if sections: final_content = prioritize_sections(sections, max_bytes) # No sections found, use raw content with truncation elif len(markdown_content.encode("utf-8")) > max_bytes: # Simple truncation encoded = markdown_content.encode("utf-8")[:max_bytes] # Handle potential multi-byte character splits while len(encoded) > 0: try: final_content = encoded.decode("utf-8") break except UnicodeDecodeError: encoded = encoded[:-1] else: final_content = "" else: final_content = markdown_content else: # No main content found final_content = f"Documentation for {service_name} is available at {docs_url}" return { "service": service_name, "content": final_content, "size_bytes": len(final_content.encode("utf-8")), "source": "gcp_docs", "docs_url": docs_url, "truncated": len(final_content.encode("utf-8")) >= max_bytes, } except httpx.HTTPStatusError as exc: if exc.response.status_code == 404: return { "service": service, "content": "", "error": "Service documentation not found", "size_bytes": 0, "source": None, } return { "service": service, "content": "", "error": f"GCP docs returned {exc.response.status_code}", "size_bytes": 0, "source": None, } except httpx.HTTPError as exc: return { "service": service, "content": "", "error": f"Failed to fetch docs: {exc}", "size_bytes": 0, "source": None, } except Exception as exc: return { "service": service, "content": "", "error": f"Failed to process docs: {exc!s}", "size_bytes": 0, "source": None, }
• src/RTFD/providers/gcp.py:661-669 (schema)

  Input/output schema defined in the tool's docstring and type annotations: service (str), max_bytes (int=20480), returns CallToolResult wrapping a dict with content, size_bytes, etc.
  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 """