{
  "markdown": "# CogniLedger MCP Server — Makuri showcase\n\nA public, read-only [Model Context Protocol](https://modelcontextprotocol.io) server operated by **CogniLedger Solutions S.R.L.** (Bucharest, Romania). It exposes structured metadata about the **Makuri** EdTech platform — 11 tools (9 info tools + 2 interactive MCP Apps panels), 3 markdown resources, and 2 guided prompts, covering mission, languages, teaching approach, pricing, safety, compliance posture, tech stack, contact channels, and free public learning resources.\n\nThis is a reference deployment demonstrating production MCP patterns under EU compliance constraints. Makuri is a High Risk AI system under EU AI Act Annex III, paragraph 3 (educational AI for minors); the v1 scope of this server is therefore deliberately narrow: **metadata only, no user data, no PII, no aggregated analytics**.\n\n- **Production endpoint:** `https://mcp.cogniledger.eu/mcp`\n- **License:** MIT\n- **Repository:** [github.com/Cogniledger/cogniledger-mcp-makuri](https://github.com/Cogniledger/cogniledger-mcp-makuri)\n- **Contact:** [leonid@cogniledger.eu](mailto:leonid@cogniledger.eu)\n\n## What this server is\n\n- Public, unauthenticated, read-only\n- 11 tools returning static metadata (no database queries against user data), including 2 interactive MCP Apps panels that render inline in supporting hosts\n- 3 markdown resources (manifesto, child-safety overview, connect guide) and 2 guided prompts\n- Designed to be called by AI assistants — Claude Desktop, Le Chat (Mistral), Cursor, ChatGPT Apps SDK, and any other MCP-capable client\n\n## What this server is not\n\n- It is not the Makuri product. End users of Makuri (children, parents) interact with [makuri.eu](https://makuri.eu), not this MCP server.\n- It does not expose user data, PII, IP addresses, behavioral analytics, or any data derived from end-user activity.\n\n## Connect\n\n### Claude Desktop\n\nEdit `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS) or `%APPDATA%\\Claude\\claude_desktop_config.json` (Windows):\n\n```json\n{\n  \"mcpServers\": {\n    \"cogniledger-makuri\": {\n      \"url\": \"https://mcp.cogniledger.eu/mcp\"\n    }\n  }\n}\n```\n\nRestart Claude Desktop. The tools appear under the connector picker.\n\n### Le Chat (Mistral)\n\nIn Le Chat settings, add a new MCP connector:\n\n- **Name:** CogniLedger — Makuri\n- **Transport:** streamable-http\n- **URL:** `https://mcp.cogniledger.eu/mcp`\n\n### Cursor\n\nEdit `~/.cursor/mcp.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"cogniledger-makuri\": {\n      \"url\": \"https://mcp.cogniledger.eu/mcp\"\n    }\n  }\n}\n```\n\n## Tools\n\n| Tool | Description |\n|---|---|\n| `get_platform_info` | Mission, target users, founding details, operating company. |\n| `get_supported_languages` | All 14 supported locales with UI / AI tutor coverage flags. Optional `locale` filter. |\n| `get_subjects` | Textbook-agnostic teaching approach, ten action buttons, learning modes. |\n| `get_pricing_tiers` | Free trial and beta subscription details. |\n| `get_safety_features` | Age gate, content filters, parental controls, AI safety guardrails. |\n| `get_compliance_matrix` | EU AI Act, GDPR, GDPR-K, COPPA, ISO 42001 — current status with disclaimer. Optional `regulation` filter. |\n| `get_tech_stack` | Frontend, backend, database, AI providers, EU data residency. |\n| `get_contact_info` | Contact channels by purpose. Optional `purpose` filter. |\n| `get_free_resources` | Free Makuri resources without registration: Slovarik vocabulary and the Romanian level test in two flavors (Quick Check, 20 questions, no email; Deep Diagnostic, 60 questions, email + certificate). |\n| `show_how_makuri_works` | Interactive MCP Apps panel explaining the Makuri learning flow and ten action buttons (`ui://makuri/how-it-works`). |\n| `show_romanian_quiz` | Interactive MCP Apps panel: Romanian mini-quiz that draws 10 random questions from a bank of 15 (levels A1–B2), RU/UK interface toggle, per-answer explanations, approximate level estimate, and CTA to the full free level test (`ui://makuri/romanian-quiz`). |\n\n### Resources\n\nThe server exposes three markdown documents via MCP `resources/list` / `resources/read`:\n\n- `makuri://docs/manifesto` — the founder-written manifesto on why Makuri exists.\n- `makuri://docs/safety-overview` — child-safety design measures (account model, data minimization, AI behavior controls).\n- `makuri://docs/connect-guide` — how to connect this server in ChatGPT, Claude, and Le Chat.\n\n(The two interactive panels above are also exposed as resources at `ui://makuri/how-it-works` and `ui://makuri/romanian-quiz`, for hosts that support MCP Apps.)\n\n### Prompts\n\nTwo guided prompts via `prompts/list`:\n\n- `evaluate_makuri_for_my_child(child_age, native_language)` — guided fit evaluation for a specific child.\n- `makuri_safety_briefing()` — honest safety briefing with explicit \"design posture, not certified compliance\" framing.\n\nFull input schemas and example responses: [`docs/TOOLS.md`](./docs/TOOLS.md). Real client transcripts: [`docs/EXAMPLES.md`](./docs/EXAMPLES.md). Compliance disclosure: [`docs/COMPLIANCE_DISCLOSURE.md`](./docs/COMPLIANCE_DISCLOSURE.md).\n\n## Local development\n\n```bash\ngit clone https://github.com/Cogniledger/cogniledger-mcp-makuri.git\ncd cogniledger-mcp-makuri\nnpm install\nnpm run dev\n```\n\nThe server starts on `http://localhost:3000`. The MCP endpoint is `http://localhost:3000/mcp`.\n\n### Smoke test\n\n```bash\n# In one terminal:\nnpm run dev\n\n# In another:\nnpm run test:smoke\n```\n\nThe smoke test connects to the local server, lists tools, calls each one, and exits non-zero on any failure.\n\n### MCP Inspector\n\n```bash\nnpx @modelcontextprotocol/inspector\n```\n\nConnect with transport `streamable-http` to `http://localhost:3000/mcp`. All 11 tools should be visible and callable, plus the 5 resources and 2 prompts.\n\n## Security and compliance\n\n- No secrets in the repository. `.env` is git-ignored; `.env.example` lists variable names only and is empty for v1.\n- No outbound API calls, no database queries, no authentication. v1 is pure static-data read.\n- Per-tool structured logging emits one JSON line per invocation with five fields only: `evt`, `tool`, `ts`, `status`, `duration_ms`. No request bodies, no IPs, no argument values.\n- See [`docs/COMPLIANCE_DISCLOSURE.md`](./docs/COMPLIANCE_DISCLOSURE.md) for the full compliance posture.\n\n## License\n\nMIT — see [`LICENSE`](./LICENSE).\n\nCopyright © 2026 CogniLedger Solutions S.R.L.\n",
  "bytes": 6399,
  "sha": "b02e4a13859edfcaa92453658e340abdb9e5a2c7621943e2bb76c51a183ff8ab",
  "repo_slug": "cogniledger/cogniledger-mcp-makuri",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_cogniledger_cogniledger_mcp_ma_25aebc78/readme"
}