Anamnesis

@oscardvs

About Anamnesis

Cross-machine memory for Claude Code: local-first, file-based agent memory that syncs across your own machines.

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "anamnesis": {
      "command": "uvx",
      "args": [
        "anamnesis-memory",
        "serve"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Anamnesis?

Anamnesis is a local-first, file-based memory layer for Claude Code that syncs automatically across all of your own machines. It captures project conventions, architecture decisions, fixes, and session summaries as plain markdown, indexed for fast retrieval and synced over your private network without any cloud account.

How to use Anamnesis?

Install by running uv tool install anamnesis-memory && anamnesis init. This registers the MCP server with Claude Code, installs session hooks, and configures the store at ~/.anamnesis. For cross‑machine sync, set up a git remote on a Tailscale mesh and run anamnesis init --remote <remote>. Claude Code gains five MCP tools: memory_search, memory_list, memory_status, memory_write, and memory_sync.

Key features of Anamnesis

File-first memory stored as human‑readable, git diff-able markdown.
Sub‑millisecond BM25 recall via a local SQLite FTS5 index.
Robust sync via git over your private Tailscale mesh (database never corrupted).
Claude‑Code‑native MCP server with read‑only auto‑query tools and session hooks.
Optional reflection loop that distills session logs into durable, provenance‑tracked notes.
A git‑like dashboard to browse, search, edit, and see memory history across machines.

Use cases of Anamnesis

Carry your coding agent’s project knowledge when switching from desktop to laptop.
Automatically capture and archive every session summary without manual steps.
Synchronize learned conventions and fixes across a fleet of personal machines.
Audit and edit your agent’s memory with full version history via the dashboard.

FAQ from Anamnesis

What does Anamnesis do?

Anamnesis provides a persistent memory layer for Claude Code that syncs across all your machines. It stores memory as markdown files, indexes them for fast retrieval, and syncs via git over your private network.

What are the prerequisites to use Anamnesis?

You need Claude Code, uv (the Astral tool installer), and git. For cross‑machine sync, you also need Tailscale installed on every machine and a shared bare git repo.

Where is my memory data stored?

All memory lives in ~/.anamnesis/memory/ as plain markdown files. A local SQLite FTS5 index is built from those files—the database is never synced, so it never corrupts.

Does Anamnesis require a cloud service?

No. Anamnesis is local‑first and syncs over your own private network via Tailscale. No cloud account or third‑party server is required.

How does cross‑machine sync work?

Sync uses a git repository over your Tailscale mesh. One always‑on machine hosts a bare git repo; each machine runs commit → pull --rebase → push. After pulling, the local SQLite index is rebuilt, making new notes searchable immediately. Conflicting edits surface as git conflicts.

Frequently asked questions

What does Anamnesis do?

What are the prerequisites to use Anamnesis?

You need Claude Code, `uv` (the Astral tool installer), and `git`. For cross‑machine sync, you also need Tailscale installed on every machine and a shared bare git repo.

Where is my memory data stored?

All memory lives in `~/.anamnesis/memory/` as plain markdown files. A local SQLite FTS5 index is built from those files—the database is never synced, so it never corrupts.

Does Anamnesis require a cloud service?

No. Anamnesis is local‑first and syncs over your own private network via Tailscale. No cloud account or third‑party server is required.

How does cross‑machine sync work?

Sync uses a git repository over your Tailscale mesh. One always‑on machine hosts a bare git repo; each machine runs `commit → pull --rebase → push`. After pulling, the local SQLite index is rebuilt, making new notes searchable immediately. Conflicting edits surface as git conflicts.

Comments

Basic information

More AI & Agents MCP servers

MCP-LLM Bridge

patruff

Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools

976

Open Multi-Agent Canvas

CopilotKit

The open-source multi-agent chat interface that lets you manage multiple agents in one dynamic conversation and add MCP servers for deep research

507

Competitor Tracker & Co.

Competitor Tracker & Co. watches your competitors' websites and tells you what changed. Every week it crawls their pricing, product, messaging and corporate pages, detects the changes that matter, and files a tagged, ran

Unreal Engine Generative AI Support Plugin

prajwalshettydev

Unreal Engine plugin for LLM/GenAI models & MCP UE5 server. OpenAI GPT-5, Deepseek R1, Claude Opus/Sonnet, Gemini 3, Grok 4, Alibaba Qwen, Kimi, ElevenLabs TTS, Inworld, OpenRouter, Groq, GLM, Ollama, Local, Meshy, Tripo, Hunyuan3D, Rodin, fal, Dashscope, Seedream. NPC AI, agenti

622

Transform MCP

Unstructured

The Unstructured Transform MCP Server connects your ADK agent to Unstructured, a document processing platform that turns raw files into structured, AI-ready data. This integration gives your agent the ability to parse PD