{
  "slug": "railway-mcp-adds-an-experimental-local-mcp-server-c5f079468769",
  "title": "railway mcp adds an experimental local MCP server",
  "tools": [
    "railway"
  ],
  "tags": [
    "agent"
  ],
  "introduced_at": "2026-03-18T06:26:05.000Z",
  "announced_at": "2026-03-18T06:26:05.000Z",
  "effective_at": null,
  "date_basis": "source-publication",
  "status": "unknown",
  "how": {
    "incantation": "If the repository coding task requires managing Railway projects, start the experimental local MCP server with `railway mcp`; it serves MCP over stdio. Use the tools exposed by that server for supported Railway operations.",
    "examples": [
      "railway mcp",
      "{\"mcpServers\":{\"railway\":{\"command\":\"railway\",\"args\":[\"mcp\"]}}}"
    ],
    "gotchas": [
      "Version: v4.32.0",
      "Availability: Experimental and subject to change.",
      "Prerequisite: The command initializes Railway CLI configuration and an authorized GraphQL client; authenticate the CLI before starting the server. The v4.32.0 command has no command-line arguments.",
      "MCP server runs locally over stdio.",
      "Docs conflict: Current Railway CLI docs describe bare `railway mcp` as a hosted MCP proxy and use `railway mcp local` for the GraphQL-backed local server. That differs from v4.32.0, whose command starts a local stdio server; current behavior must not be substituted for the card's version."
    ]
  },
  "why_it_matters": "Determine whether the experimental Railway MCP server applies to a repository coding task that requires managing Railway projects.",
  "sources": [
    "https://github.com/railwayapp/cli/releases/tag/v4.32.0",
    "https://github.com/railwayapp/cli/pull/813",
    "https://github.com/railwayapp/cli/blob/v4.32.0/src/commands/mcp/mod.rs",
    "https://github.com/railwayapp/cli"
  ],
  "rank": 0.98062,
  "judge": {
    "publish": true,
    "judge_model": "codex-cli/gpt-6-luna+typesafe/jev-1.13.0",
    "prompt_version": "extract-v7+enrich-v2+judge-v2x3+jev-publish-v1"
  },
  "review_state": "approved",
  "provenance": {
    "run_id": "wave1b-20261016",
    "verdict_id": "wave1b-20261016:wave1b-20261016-ab3911f5f43c105a-3:gate",
    "evidence_hash": "ed8326dc8f58a4c11d1f762ac3bfba1b3662538f6aaad37824d6f277a253fd32"
  },
  "review": {
    "draft_hash": "1355fbf2c065f2c7218911ef68547b89223c18d8353ce6132b6f0b6f40534995",
    "reviewer": "publication-gate",
    "reviewed_at": "2026-10-07T00:00:00.000Z",
    "decision": "wave1b-20261016:gate",
    "reason": "3 of 3 judge runs approved with p_yes >= 0.9 (mean 0.98 >= 0.9; 2 needed) and Jev gave 0.75 (>= 0.6 needed)."
  }
}