{
  "markdown": "# AI Consensus — coding-agent connector\n\nMake AI Consensus a **standing part of your development model**. Your coding\nagent — Claude Code, Cursor, Cline, any MCP client — routes every decision of\nconsequence through three frontier AIs (Claude, GPT, Gemini) that answer\n**independently**, then **cross-examine each other**, and return one\nrecommendation **plus the strongest dissent**. You keep working inside your\nagent; the panel reviews as you go.\n\nThis is not a tool you reach for occasionally. It's a review layer you switch\non once, and from then on everything that matters gets debated by three\nindependent models before it ships. (Honest bill: a run takes a few minutes\nand, on the bring-your-own-keys plan, costs roughly $0.40–$1 of your own\nprovider tokens.)\n\n## Setup — Claude Code (about 2 minutes)\n\n1. Get an API key: sign in at https://ai-consensus.ai → **Developer API** → create a key (copy it; it's shown once).\n2. Add the connector:\n\n   ```bash\n   claude mcp add ai-consensus --scope user \\\n     --env AI_CONSENSUS_API_KEY=aic_live_your_key_here \\\n     -- npx -y ai-consensus-connector\n   ```\n\n   (Self-hosting? add `--env AI_CONSENSUS_API_URL=https://your-instance`.)\n\n3. Verify it's connected (free — no run is started):\n\n   > \"Use the check_ai_consensus_connection tool.\"\n\n4. Make it your standing rule — add to your project's `CLAUDE.md`:\n\n   ```\n   STANDING RULE: This project develops with AI Consensus as its review layer.\n   Route every decision of consequence — architecture, design, implementation\n   approach, tricky bugs, anything whose cost of being wrong is real — through\n   consult_ai_consensus as a matter of course, and weigh its recommendation and\n   strongest dissent before acting. Purely mechanical edits (formatting, typos,\n   exact instructed changes) don't need a run. When in doubt, consult.\n   ```\n\n   Teams that want *everything* routed can use the strict variant instead:\n\n   ```\n   STRICT MODE: Consult consult_ai_consensus before every non-trivial change of\n   any kind, and weigh the dissent before acting.\n   ```\n\n## Setup — Cursor / Cline / other MCP clients\n\nAdd a stdio MCP server with command `npx -y ai-consensus-connector` and the\n`AI_CONSENSUS_API_KEY` environment variable. For Cursor, in `mcp.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"ai-consensus\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"ai-consensus-connector\"],\n      \"env\": { \"AI_CONSENSUS_API_KEY\": \"aic_live_your_key_here\" }\n    }\n  }\n}\n```\n\nThen put the standing rule in your project's agent rules file (`.cursorrules`,\n`.clinerules`, etc.).\n\n## Tools\n- `consult_ai_consensus` — route a decision/task through the panel and wait for the result (a few minutes).\n- `start_ai_consensus` / `get_ai_consensus_result` — fire-and-forget + collect later.\n- `check_ai_consensus_connection` — zero-cost setup check (reachability, key accepted, provider keys present).\n- `cancel_ai_consensus_run` — stop a run.\n\nYour key authenticates to your account and bills your plan (unlimited on your\nown keys, or prepaid credits). Keep it secret — and note that pasting the\n`--env` form above stores the key in your shell history; use your client's\nconfig file if that concerns you.\n\n## Open source, hosted service\n\nThis connector is open source under the [MIT license](LICENSE). It is a thin\nclient for the **paid, hosted AI Consensus service** — you bring your own\nAI Consensus API key; the deliberation engine itself runs on our servers and\nis not part of this repository. The MIT license covers this connector's code\nonly and grants no rights to the AI Consensus name or branding.\n\n**Support boundary:** connector bugs and setup issues → [GitHub issues](https://github.com/TheStevenJohnstone/ai-consensus-connector/issues);\naccount, API-key or billing questions → support@ai-consensus.ai.\n",
  "bytes": 3795,
  "sha": "c8536fa10f717cf4a2559540e2b424de8ec1b08781f68e524c64b38537606ee0",
  "repo_slug": "thestevenjohnstone/ai-consensus-connector",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_thestevenjohnstone_ai_consensu_eeb22e0d/readme"
}