{
  "markdown": "# UnRemind.me MCP Server\n\nRemote [MCP](https://modelcontextprotocol.io) server for\n**[UnRemind.me](https://unremind.me)** — context-aware reminders that surface\nwhen your *situation* matches, instead of firing at a fixed time.\n\nAn \"Unreminder\" carries context tags — where you are, what device you have,\nwhether you're online, who's around, how much time and energy you have — and\nsurfaces when your current context matches. Tasks can also be **delegated to\nAI agents**, each with scoped access rights and an audit log the owner\nreviews.\n\n[![smithery badge](https://smithery.ai/badge/submissions-epan/unremind-me)](https://smithery.ai/servers/submissions-epan/unremind-me)\n[![smithery badge](https://smithery.ai/badge/submissions-epan/unremind-me-submit)](https://smithery.ai/servers/submissions-epan/unremind-me-submit)\n\n**Endpoint** · `https://unremind.me/mcp`\n**Transport** · Streamable HTTP (JSON-RPC 2.0) · **Auth** · Bearer token ·\n**Tools** · 16 · **Prompts** · 4\n\n---\n\n## Connect\n\n1. Install UnRemind.me ([Android](https://unremind.me/) or the\n   [web app](https://app.unremind.me/)) and sign in.\n2. In the app: **Settings → AI agents → MCP access → Create token.** The token\n   is shown once.\n3. Add the server to your MCP client:\n\n```json\n{\n  \"mcpServers\": {\n    \"unremind\": {\n      \"type\": \"streamable-http\",\n      \"url\": \"https://unremind.me/mcp\",\n      \"headers\": { \"Authorization\": \"Bearer unr_YOUR_TOKEN\" }\n    }\n  }\n}\n```\n\nVerified with Claude and ChatGPT custom connectors (desktop and mobile).\nFull walkthrough: **<https://unremind.me/mcp/>**\n\nEach token is a separate *agent* in the app, with its own access rights, its\nown audit log, and an optional expiry. Revoke any of them from the same\nscreen without affecting the others.\n\n## Tools\n\nSee **[TOOLS.md](TOOLS.md)** for all 16 with descriptions. Every tool carries\nMCP behavioural annotations (`readOnlyHint`, `destructiveHint`,\n`idempotentHint`, `openWorldHint`) so a client knows which are safe to call\nwithout confirming and which must not be blindly retried.\n\n## Prompts\n\nFour user-invocable prompts, available from your MCP client's prompt menu:\n\n| Prompt | |\n|---|---|\n| **What can I do right now?** | Give it your available minutes and what you have with you; it finds the tasks that actually fit |\n| **Plan my day** | Reviews overdue/due-soon work and checks real calendar conflicts before proposing times |\n| **Capture this as an Unreminder** | Turns raw text into a well-tagged task |\n| **Resume where I left off** | For an agent starting a session: orient, then continue instead of starting cold |\n\n## Where to start\n\n`orient` is the one to call first — it returns who the user is, how they want\nto be worked with, what you were last doing, and what needs attention. One\ncall instead of guessing.\n\n## Discovery documents\n\n| | |\n|---|---|\n| Capability manifest | <https://unremind.me/.well-known/ai-capabilities.json> |\n| Auth metadata (RFC 9728) | <https://unremind.me/.well-known/oauth-protected-resource> |\n| LLM-facing site map | <https://unremind.me/llms.txt> |\n| Health / liveness | <https://unremind.me/health> |\n\nA `401` from the endpoint carries\n`WWW-Authenticate: Bearer resource_metadata=\"…\"`, so a compliant client can\ndiscover the auth model without being told.\n\n## Deep links\n\nEvery entity has both an app link and an https twin:\n\n```\nunremind://u/{id}              # opens the app\nhttps://unremind.me/u/{id}     # always resolves\n```\n\n**Print both** when referring a person to something — you can't tell whether\nthey're reading on the device that holds the app or on a desktop where a\ncustom scheme resolves to nothing.\n\n## Design notes for agent authors\n\n- **`create_unreminder` submits for approval**; it does not silently appear in\n  the user's list. Writes that a person will act on go through them.\n- **Context tags describe what a task *requires*** to be doable (`Computer`\n  means the task needs a computer) — never what the user currently has.\n- **`save_progress` notes are yours, not a source of truth.** The user edits\n  tasks in the app between sessions; always re-read the live task before\n  acting on a recollection.\n\n## Status\n\nProduction. Server `unremind-mcp` v2.0.0. Liveness at\n<https://unremind.me/health> (`200` healthy / `503` degraded), checked\ncontinuously.\n\nIssues and questions: [open an issue](https://github.com/esfih/unremind-me-mcp/issues)\nor <mailto:support@unremind.me>.\n\n## License\n\n[MIT](LICENSE) for this repository's contents. The UnRemind.me service itself is\ngoverned by its [Terms](https://unremind.me/terms) and\n[Privacy Policy](https://unremind.me/privacy).\n\n## Two servers\n\nUnRemind.me exposes two MCP servers with different auth models. They\ncross-reference each other, so an agent that finds either learns of the other.\n\n| | Endpoint | Auth | For |\n|---|---|---|---|\n| **UnRemind.me** | `https://unremind.me/mcp` | bearer token | Manage *your own* reminders — 16 tools, 4 prompts |\n| **UnRemind.me Submit** | `https://unremind.me/mcp/submit` | **none** | Send a reminder *to someone else* with a public page — 2 tools, 1 prompt |\n\n## Listed on\n\n- [Smithery — UnRemind.me](https://smithery.ai/servers/submissions-epan/unremind-me)\n- [Smithery — UnRemind.me Submit](https://smithery.ai/servers/submissions-epan/unremind-me-submit)\n",
  "bytes": 5269,
  "sha": "59b4eedd558b1b746c395f13b9d55e317adf011a7ffdbbf2c083db4b6bda5c82",
  "repo_slug": "esfih/unremind-me-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_esfih_unremind_me_61fb2acd/readme"
}