rsperformance-online-mcp.yml
RS Performance advertises two remote MCP servers on the host mcp.rs3d.pl through its own /.well-known/mcp.json. The first, "Diagnosta RS" at https://mcp.rs3d.pl/, is a live FastMCP Streamable HTTP server that answered an anonymous initialize + tools/list on 2026-09-19 with 14 tools, 3 resources and 0 prompts. The second, a Laravel MCP server at https://mcp.rs3d.pl/laravel/mcp/rs-knowledge, is advertised in mcp.json, ai-plugin.json, agents.json and the gateway freshness beacon but returned HTTP 500 "Server Configuration Error: MCP_SERVER_TOKEN is missing" to initialize, tools/list and GET at probe time.
One-click install for Cursor, VS Code, Claude, and 20+ other MCP clients, powered by API Commons MCP Install — visit install.apicommons.org for more information.
Endpoint
Hosted endpoint · transport streamable-http
Connect
This is a remote MCP server — point an MCP client at the endpoint URL. Clients with native remote support (Claude, Cursor, VS Code, …):
For stdio-only clients (older Claude Desktop, etc.), bridge with mcp-remote:
If the server requires authentication, add the provider's token/header (e.g. an X-API-Key) per its docs. A quick reachability check:
Tool quality observed · not scored
tools with readOnly / destructive hints
parameters described (35 total)
state-changing tools without destructiveHint
tool-list tokens (14 tools)
tools over ~1,200 tokens
Unhinted state-changing: write_intake_note
From an anonymous initialize + tools/list capture measured 2026-10-20; tokens are chars/3.6 of each tool definition. Recorded for the Kin Score's agent readiness as an observed signal — it contributes no points yet.
Tools
semantic_search— Search the RS Performance knowledge base using AI semantic search. Returns the most relevant documents about services, FAQ, business info, blog posts, and DTC codes. Use doc_type to filter: 'service', 'blog', 'dtc', 'business', 'faq', or '' for all.diagnostic_search— Compatibility bridge for legacy Diagnosta clients expecting diagnostic_search. For end-user replies, prefer `assistant_answer`, `confidence`, `recommended_next_steps`, and `ask_back`. Use raw `hits` only as evidence/debug, not as the main final response.diagnostic_web_research— Search the live web through local SearXNG and extract readable evidence with Trafilatura. Prefer this when the query needs fresh public context or forum/TSB-style evidence from the internet. For end-user replies, prefer `assistant_answer`, `confidence`, `recommended_next_steps`, and `ask_back`. Use raw `hits` only as evidence/debug.diagnostic_search_live— Combine local Diagnosta knowledge with live web research. Uses canonical DTC JSON, private notes, Qdrant retrieval, local SearXNG, and Trafilatura extraction. Prefer this for owner-only deep diagnosis when local knowledge should be augmented by current public web sources.diagnostic_deep_research— Run owner-only deep internet research for Diagnosta. Uses local SearXNG discovery with a separate deep-ingest worker that can enrich from HTML pages, documents, and YouTube transcripts. Prefer this when you explicitly want broader internet evidence beyond the standard live web search.diagnostic_ingest_brand_knowledge— Collect and save a brand knowledge packet on the VPS for private Diagnosta use. Use this when the owner says things like "pobierz dane o Fiat" or asks to gather a broader research pack for a manufacturer. This tool does not promise "the whole internet". It runs a bounded multi-query scrape, saves artifacts on the VPS, and returns the saved paths. `topics_csv` is optional and can narrow the batch, e.g. `Multijet,Body Computer,CAN`.health— Return the current health and directory layout of the Diagnosta RS MCP service.list_sources— List the top-level directories and searchable files inside the Diagnosta knowledge root.search_notes— Search the Diagnosta knowledge base for a phrase and return matching file snippets.read_note— Read a single Diagnosta knowledge document from the private data store.write_intake_note— Write a new intake note into the Diagnosta knowledge base.repair_reports_overview— Fetch an overview of RS Performance repair reports from the secured Laravel API.search_repair_reports— Search RS Performance repair reports by phrase, vehicle data, fault code or service.get_repair_report— Fetch one repair report by numeric ID or slug from the secured Laravel API.
About MCP
The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.
Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.
Work with this as data
Every MCP server here is available over the APIs.io API and to AI agents over MCP.