{
  "markdown": "# RCA-MCP Connector\n\n<!-- mcp-name: io.github.dave1362/rca-mcp-connector -->\n\n![Python](https://img.shields.io/badge/python-3.12+-blue) ![License](https://img.shields.io/badge/license-MIT-green) ![PulseMCP](https://img.shields.io/badge/PulseMCP-listed-orange)\n\n> **⚠️ Early access.** RCA-MCP's backend is live and this connector has been\n> verified end-to-end against it. The `api.rca-mcp.com` custom domain isn't\n> wired up yet — point `RCA_MCP_API_URL` at the current backend URL below.\n> Tool schemas and documentation may still change before the first stable\n> public launch.\n\n## What is RCA-MCP?\n\nThe only MCP server purpose-built for causal Root Cause Analysis. 56 tools covering\ncausal graph construction, 10 RCA model families plus 3 dedicated PyRCA algorithms\n(Salesforce PyRCA, BSD-3-Clause), multi-model consensus, and PDF/HTML/Excel/Markdown\nreport generation. Works with Claude, Ollama, Groq, OpenAI, Gemini, LangChain,\nCursor — 10 providers.\n\n---\n\n## Quick Start (2 minutes)\n\n`rca-mcp-connector` is a published PyPI package — no clone needed. Point any MCP\nclient at it with `uvx` (or `pip install rca-mcp-connector` if you'd rather manage\nthe install yourself):\n\n```bash\nuvx rca-mcp-connector\n```\n\nGet a free API key at [rca-mcp.pages.dev](https://rca-mcp.pages.dev) — no credit\ncard required — then set `RCA_MCP_API_KEY` in your MCP client's config (examples\nbelow).\n\n---\n\n## Claude Code Setup\n\nAdd to `.mcp.json` in your workspace root:\n\n```json\n{\n  \"mcpServers\": {\n    \"rca-mcp\": {\n      \"command\": \"uvx\",\n      \"args\": [\"rca-mcp-connector\"],\n      \"env\": {\n        \"RCA_MCP_API_URL\": \"https://rcamcp-production.up.railway.app\",\n        \"RCA_MCP_API_KEY\": \"your_api_key_here\"\n      }\n    }\n  }\n}\n```\n\n## Ollama Setup\n\n```bash\ngo install github.com/mark3labs/mcphost@latest\nmcphost -m ollama:qwen3:14b --config providers/mcp-servers.json\n```\n\n## OpenAI Agents SDK\n\n```python\nfrom agents import Agent, MCPServerStdio\nimport asyncio\n\nasync def main():\n    async with MCPServerStdio(\n        params={\n            \"command\": \"uvx\",\n            \"args\": [\"rca-mcp-connector\"],\n            \"env\": {\n                \"RCA_MCP_API_URL\": \"https://rcamcp-production.up.railway.app\",\n                \"RCA_MCP_API_KEY\": \"your_api_key_here\",\n            },\n        }\n    ) as rca_server:\n        agent = Agent(name=\"RCA Agent\", model=\"gpt-4o\", mcp_servers=[rca_server])\n        result = await agent.run(\"Find the root cause of the API latency spike.\")\n        print(result.final_output)\n\nasyncio.run(main())\n```\n\n## LangChain\n\n```python\nfrom langchain_mcp_adapters.client import MultiServerMCPClient\nfrom langchain_anthropic import ChatAnthropic\nfrom langgraph.prebuilt import create_react_agent\nimport asyncio\n\nasync def main():\n    async with MultiServerMCPClient({\n        \"rca-mcp\": {\n            \"command\": \"uvx\", \"args\": [\"rca-mcp-connector\"],\n            \"env\": {\n                \"RCA_MCP_API_URL\": \"https://rcamcp-production.up.railway.app\",\n                \"RCA_MCP_API_KEY\": \"your_api_key_here\",\n            },\n            \"transport\": \"stdio\",\n        }\n    }) as client:\n        tools = await client.get_tools()\n        agent = create_react_agent(ChatAnthropic(model=\"claude-sonnet-4-6\"), tools)\n        result = await agent.ainvoke({\"messages\": [{\"role\": \"user\", \"content\": \"Run an FMEA analysis\"}]})\n        print(result[\"messages\"][-1].content)\n\nasyncio.run(main())\n```\n\nSee `providers/` for ready-to-use config templates and full examples (Groq, Gemini,\nOpenRouter, Claude Desktop).\n\n---\n\n## Third-Party Licences\n\n**PyRCA (Salesforce):** BSD-3-Clause\nCopyright (c) 2022, salesforce.com, inc.\nhttps://github.com/salesforce/PyRCA\n\nAlgorithms in `rca_pyrca_*` tools are independently-written adaptations of PyRCA's\npublished methods (Zheng et al. 2023, arXiv:2306.11417), not direct copies of PyRCA\nsource code, per the private API's `models/pyrca_adapter.py`.\n\n---\n\n## Citing RCA-MCP\n\n```bibtex\n@software{rcamcp2026,\n  title  = {RCA-MCP: An MCP Server for Causal Root Cause Analysis},\n  author = {davetj},\n  year   = {2026},\n  url    = {https://github.com/dave1362/rca-mcp-connector},\n  note   = {v4.1.14}\n}\n```\n",
  "bytes": 4128,
  "sha": "33afdd62aa4adc6ce68b3e53eeb63ea8262659751e2caa0ab5cd688d7de23e9a",
  "repo_slug": "dave1362/rca-mcp-connector",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_dave1362_rca_mcp_connector_1745bbbe/readme"
}