{
  "markdown": "# TwoThumbs MCP\n\n**Verify what your AI agent actually did — not what it says it did.**\n\nTwoThumbs is a hosted, remote [MCP](https://modelcontextprotocol.io) server built by NOMOI Labs.\nIt engineers an evidence-backed prompt, challenges weak premises, then checks the real artifact\nand downstream outcome a public web surface actually produced — instead of trusting a\nself-reported \"success\" message. It also carries a much larger internal prompt-engineering and\nverification-lifecycle tool set (51 tools total), of which the 3 tools below are exposed publicly\nover the hosted HTTP endpoint.\n\nThis repository is a thin discovery wrapper: an MCP client config, a Claude Code plugin manifest,\nand this README. There is no source code to install or build — the server itself is hosted and\nrun by NOMOI at [twothumbs.co](https://twothumbs.co).\n\n## Connect\n\nThe server speaks Streamable HTTP at:\n\n```\nhttps://twothumbs.co/mcp\n```\n\nNo local install, no package, no API key required to use the anonymous tools.\n\n### Claude Code\n\n```bash\nclaude mcp add --transport http twothumbs https://twothumbs.co/mcp\n```\n\nOr install this repository as a Claude Code plugin (adds the same server via `.mcp.json`):\n\n```\n/plugin marketplace add asselwan/twothumbs-mcp\n/plugin install twothumbs@twothumbs-mcp\n```\n\n### Cursor\n\nAdd to your MCP settings (`Cursor Settings → MCP`):\n\n```json\n{\n  \"mcpServers\": {\n    \"twothumbs\": {\n      \"url\": \"https://twothumbs.co/mcp\"\n    }\n  }\n}\n```\n\n### Generic MCP client\n\n```json\n{\n  \"mcpServers\": {\n    \"twothumbs\": {\n      \"type\": \"http\",\n      \"url\": \"https://twothumbs.co/mcp\"\n    }\n  }\n}\n```\n\nWorks with any client that supports the Streamable HTTP transport — Claude Code, Cursor, Codex,\nand CI/headless clients included.\n\n## Tools\n\nAvailable anonymously, with a free daily quota:\n\n| Tool | What it does |\n|---|---|\n| `run_verdict` | Exercises every visible affordance on a public URL — clicking, submitting forms with clearly marked synthetic data — then returns a pass/fail verdict grounded in the real downstream outcome, not a self-reported success message. |\n| `list_surfaces` | Lists the target surfaces currently configured for `run_verdict`. |\n| `redeem_code` | Exchanges an access code for an API key. |\n\nAn API key additionally unlocks `get_last_verdict`, `get_balance`, and `set_auto_topup`.\n\n## Pricing\n\n- **Free** — 3 surface scans/day, no signup required.\n- **$9** — 100 credits.\n- **$29** — 500 credits.\n\nBilling is invoice-based; email through [twothumbs.co](https://twothumbs.co) to request a key, or\nuse `redeem_code` with an access code.\n\n## Links\n\n- Product: [twothumbs.co](https://twothumbs.co)\n- Docs / quickstart: [twothumbs.co/docs.html](https://twothumbs.co/docs.html)\n- Privacy policy: [twothumbs.co/privacy](https://twothumbs.co/privacy)\n\n## License\n\nThe contents of this repository (plugin manifest, client config, documentation) are MIT licensed\n— see [LICENSE](./LICENSE). This license covers only this wrapper repository; it does not extend\nto the TwoThumbs product or hosted service, which remain the property of NOMOI.\n",
  "bytes": 3068,
  "sha": "23c10fc3297fb0ceb84d99671f3ba1daf1692c2a8dcc4205f1bae0a068af3976",
  "repo_slug": "asselwan/twothumbs-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_asselwan_twothumbs_74dd0530/readme"
}