Anamnese
Give Claude Code or Cursor a persistent memory layer so notes, tasks, goals, and context survive across chats and tools.
Overview
Anamnese is an MCP server for the Build phase that provides a self-improving memory layer for notes, tasks, and goals across agent sessions.
What is this MCP server?
- Remote streamable-http MCP at anamneseai.app—no local npm package required for the hosted endpoint
- Self-improving memory for notes, tasks, and goals referenced by the agent over time
- Designed to be remembered everywhere the same MCP client connects
- Solo-builder fit: one account backs personal context instead of re-pasting project state each session
- Protocol: streamable-http remote at https://anamneseai.app/api/mcp
- Server schema version referenced: MCP server.schema.json (2025-12-11)
- Packaged server version: 1.0.0
What problem does it solve?
Agents forget project context, personal goals, and open tasks every time you start a new chat or switch machines.
Who is it for?
Solo builders who live in Claude Code or Cursor and want one persistent memory surface without building custom vector storage first.
Skip if: Teams that need audited, role-based knowledge bases with strict compliance boundaries—use a formal KMS or DB-backed RAG instead.
What do I get? / Deliverables
After connecting the remote MCP endpoint, your agent can read and update a shared memory layer so recurring work picks up where you left off.
- Registered remote MCP connection to Anamnese
- Agent-accessible memory for notes, tasks, and goals
- Cross-session context recall during coding workflows
Recommended MCP Servers
Journey fit
Agent tooling is the primary shelf because the server is wired into MCP clients during product build and daily agent workflows. agent-tooling fits MCP registration, remote streamable-http endpoints, and skills that need durable user memory outside a single session.
How it compares
MCP memory integration, not an in-repo agent skill or local SQLite notes hack.
Common Questions / FAQ
Who is app.anamneseai/anamnese for?
It is for indie developers and solo founders using MCP agents who want durable notes, tasks, and goals outside a single conversation.
When should I use app.anamneseai/anamnese?
Use it when you are wiring agent-tooling during a build and tired of re-explaining project state, or when you iterate in operate and need continuity across sessions.
How do I add app.anamneseai/anamnese to my agent?
Register the streamable-http remote URL https://anamneseai.app/api/mcp in your MCP client config per anamneseai.app setup, then restart or reload the agent.