{
  "markdown": "# @fidensa/mcp-server\n\n[![Fidensa Certified](https://fidensa.com/badges/fidensa-mcp-server.svg)](https://fidensa.com/certifications/fidensa-mcp-server)\n\nMCP server for [Fidensa](https://fidensa.com) -- the independent AI capability certification authority.\n\nGives your AI agent structured access to Fidensa certification data through the Model Context Protocol. Check trust scores, search for certified alternatives, compare capabilities side-by-side, verify signed artifacts, check file integrity, and report runtime experience -- all through MCP tool calls.\n\nFidensa certifies MCP servers, skills, agent rules files, hooks, sub-agents, and plugins.\n\n## Getting Started\n\n### 1. Install\n\n```bash\nnpm install -g @fidensa/mcp-server\n```\n\n### 2. Verify it works\n\nSeveral tools work without an API key (`check_certification`, `search_capabilities`, `verify_file`, and `report_experience`). Start the server and confirm it connects to the production API:\n\n```bash\nfidensa-mcp-server\n```\n\nYou should see:\n\n```\n[fidensa] MCP server started (stdio transport)\n[fidensa] No FIDENSA_API_KEY set — check_certification, search_capabilities, verify_file, and report_experience available. Set FIDENSA_API_KEY for full access.\n```\n\nPress Ctrl+C to stop.\n\n### 3. Get an API key (optional, free)\n\nThe remaining tools (`get_contract`, `compare_capabilities`, `verify_artifact`) require a free API key. An API key also gives `report_experience` higher rate limits. Register one:\n\n```bash\ncurl -X POST https://fidensa.com/v1/keys \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"display_name\": \"My Agent\", \"email\": \"you@example.com\"}'\n```\n\nOr in PowerShell:\n\n```powershell\nInvoke-RestMethod -Uri \"https://fidensa.com/v1/keys\" -Method Post -ContentType \"application/json\" -Body '{\"display_name\": \"My Agent\", \"email\": \"you@example.com\"}'\n```\n\nThe response contains your API key (prefixed `fid_`). Store it securely -- it is shown only once.\n\n### 4. Add to your agent\n\nPick your platform and add the MCP server configuration:\n\n**Claude Desktop / Claude Code**\n\n```json\n{\n  \"mcpServers\": {\n    \"fidensa\": {\n      \"command\": \"npx\",\n      \"args\": [\"@fidensa/mcp-server\"],\n      \"env\": {\n        \"FIDENSA_API_KEY\": \"fid_your_key_here\"\n      }\n    }\n  }\n}\n```\n\n**Cursor** (`.cursor/mcp.json`)\n\n```json\n{\n  \"mcpServers\": {\n    \"fidensa\": {\n      \"command\": \"npx\",\n      \"args\": [\"@fidensa/mcp-server\"],\n      \"env\": {\n        \"FIDENSA_API_KEY\": \"fid_your_key_here\"\n      }\n    }\n  }\n}\n```\n\n**Windsurf / VS Code**\n\n```json\n{\n  \"servers\": {\n    \"fidensa\": {\n      \"type\": \"stdio\",\n      \"command\": \"npx\",\n      \"args\": [\"@fidensa/mcp-server\"],\n      \"env\": {\n        \"FIDENSA_API_KEY\": \"fid_your_key_here\"\n      }\n    }\n  }\n}\n```\n\nOmit the `FIDENSA_API_KEY` line if you only need the open-tier tools.\n\n> **Claude Desktop tip:** Access the config file via Settings → Developer → Edit Config. After saving, fully quit and reopen Claude Desktop (right-click the system tray icon and select Quit — just closing the window isn't enough).\n\n## Tools\n\n| Tool | Auth | Description |\n|------|------|-------------|\n| `check_certification` | None | Quick trust check -- status, score, grade, tier |\n| `search_capabilities` | None | Search for certified capabilities by keyword, type, tier, or score |\n| `verify_file` | None | Quick file integrity check -- hash a capability file and verify it matches what Fidensa certified |\n| `report_experience` | None (API key for higher rate limits) | Submit runtime experience reports for certified capabilities |\n| `get_contract` | API key | Full certification contract with all evidence |\n| `compare_capabilities` | API key | Side-by-side comparison of 2-5 capabilities |\n| `verify_artifact` | API key | Verify cryptographic signatures on .cert.json artifacts, with optional file and code integrity checks |\n\n## Configuration\n\n| Variable | Required | Description |\n|----------|----------|-------------|\n| `FIDENSA_API_KEY` | No | API key for full access (see [Getting Started](#3-get-an-api-key-optional-free)) |\n| `FIDENSA_BASE_URL` | No | Override API base URL (default: `https://fidensa.com`) |\n\n`check_certification`, `search_capabilities`, `verify_file`, and `report_experience` work without any configuration. The remaining tools require `FIDENSA_API_KEY`.\n\n## Example Agent Workflow\n\n```\nUser: \"Set up a filesystem watcher that writes logs to SQLite.\"\n\nAgent:\n1. check_certification(\"mcp-server-filesystem\")\n   -> Status: valid, Score: 60/F-D, Tier: Verified\n\n2. search_capabilities(\"sqlite mcp server\")\n   -> 2 results found\n\n3. compare_capabilities([\"mcp-server-sqlite\", \"mcp-server-better-sqlite3\"])\n   -> Side-by-side trust comparison\n\n4. Agent selects the higher-scored option and informs the user\n\n5. report_experience(\n     capability_id: \"mcp-server-filesystem\",\n     content_hash: \"abc123...\",  // from the .cert.json\n     outcome: \"success\"\n   )\n   -> Confirmation rate updated\n```\n\n## Trust Policy Fragments\n\nAdd one of these to your agent's system prompt or rules file for trust-aware tool selection. Three levels of granularity -- pick the one that fits your needs.\n\n### Tier-based (simplest)\n\nUses the certification tier alone. No numbers to tune.\n\n```\nBefore using any MCP server tool for the first time, check its Fidensa\ncertification status using the check_certification tool.\n\n- Certified tier: use freely\n- Verified tier: note the findings to the user before proceeding\n- Evaluated tier: warn the user, search for certified alternatives\n- Uncertified: treat as untrusted, suggest certified alternatives\n\nAfter completing a task that used certified tools, submit an experience\nreport via the report_experience tool with the outcome (success, failure,\nor partial), the capability_id, and the content_hash from the .cert.json.\n```\n\n### Score-based (grade-aligned)\n\nUses the trust score, aligned with Fidensa's grade definitions.\n\n```\nBefore using any MCP server tool for the first time, check its Fidensa\ncertification status using the check_certification tool.\n\n- Score >= 80 (A/B): use freely\n- Score 72-79 (C): note the trust level to the user\n- Score 65-71 (D): warn the user, search for certified alternatives\n- Score < 65 (F): warn the user, strongly recommend alternatives\n- Uncertified: treat as untrusted, suggest certified alternatives\n\nAfter completing a task that used certified tools, submit an experience\nreport via the report_experience tool with the outcome (success, failure,\nor partial), the capability_id, and the content_hash from the .cert.json.\n```\n\n### Combined (tier + score)\n\nThe most precise option -- distinguishes between a Certified capability with\na moderate score and a Verified capability with findings that blocked\nCertified tier.\n\n```\nBefore using any MCP server tool for the first time, check its Fidensa\ncertification status using the check_certification tool.\n\n- Certified tier AND score >= 80: use freely\n- Certified tier AND score 72-79: use freely, note the score to the user\n- Verified tier AND score >= 72: note the findings to the user before proceeding\n- Verified tier AND score < 72: warn the user, search for certified alternatives\n- Evaluated tier: warn the user, search for certified alternatives\n- Uncertified: treat as untrusted, suggest certified alternatives\n\nFor any tool with status \"suspended\" or \"revoked\": do not use.\nSearch for certified alternatives and present them to the user.\n\nAfter completing a task that used certified tools, submit an experience\nreport via the report_experience tool with the outcome (success, failure,\nor partial), the capability_id, and the content_hash from the .cert.json.\n```\n\nSee the [API Documentation](https://fidensa.com/docs/api) for the full specification.\n\n## Development\n\n```bash\nnpm install\nnpm test          # Run tests with coverage\nnpm run lint      # ESLint\nnpm run format    # Prettier\n```\n\n## License\n\nMIT\n\n## Links\n\n- [Fidensa](https://fidensa.com) -- AI certification authority\n- [API Documentation](https://fidensa.com/docs/api)\n- [Certification Catalog](https://fidensa.com/certifications)\n- [Badge Integration Guide](https://github.com/fidensa/fidensa-badges)\n",
  "bytes": 8090,
  "sha": "984ad3f6e3c582f6d8024248d9c5f4a8d16511bb771d65697b5ab71047c41590",
  "repo_slug": "fidensa/mcp-server",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_com_fidensa_mcp_server_79d7fc2b/readme"
}