a-memory

Your AI agents forget. a-memory makes them remember. 4-tier agent memory with hybrid search, a real knowledge graph, and envelope encryption — all in plain SQLite files. Zero cloud. Zero external APIs.

Naming note: the product name is a-memory; this repository is still Cipher208/a-memory (renaming is pending — all links above use the current name).

Why SQLite?

Every other memory server sends your agent's data through a cloud API or requires a separate vector database.

a-memory stores everything in SQLite files on your machine.

• Zero infrastructure. No Docker, no database server, no embedding API keys.
• Zero data leaving your network. Works air-gapped.
• Layer-isolated by design. User facts and agent identity never share a namespace.
• One directory = entire memory. Back up with cp, sync with rsync.

Why this exists

Three problems a-memory solves:

① Agent self-evolution — your AI stops repeating mistakes between sessions. It remembers decisions, errors, and corrections in a dedicated agent layer, and an hourly consolidation sweep promotes what matters into long-term facts.

② User persona persistence — your agent knows who it's talking to even after weeks of silence. Preferences, history, emotional context live in the user layer, isolated from agent identity.

③ Project continuity — project tracks per-project context: decisions with rationale and outcomes, artifact maps, a graphify-powered code index — so a fresh session picks up where the last one left off.

Get started

git clone https://github.com/Cipher208/a-memory.git
cd a-memory
uv sync
uv run ariel-memory          # MCP server on stdio — connect from any MCP client

Point your MCP client at it:

{
  "mcpServers": {
    "a-memory": {
      "command": "/path/to/a-memory/.venv/bin/python",
      "args": ["mcp_server/server.py", "--transport", "stdio"]
    }
  }
}

HTTP transport with dashboard:

uv run ariel-memory --transport http --port 8000 --dashboard

PyPI package (pip install a-memory) ships with the repo rename. See Roadmap.

The five primitives

Agents see exactly five tools — one verb per intent, no tool-choice paralysis:

Quick demo — Python MCP client:

# think — routed to the right store automatically
await session.call_tool("think", {"text": "User prefers dark mode", "layer": "user"})

# dream — finds it across every store, a week later
res = await session.call_tool("dream", {"query": "dark mode preference"})
print(res["summary"])

~30 additional fine-grained operations (typed CRUD per store, sessions, ops/admin) stay available behind ARIEL_EXPOSE=all.

Features

Architecture

graph TD
    A[LLM Agent] -->|MCP Protocol| B[mcp_server]
    B --> C{Importance Scoring}
    C --> D[L1: ReflexBuffer]
    D --> E[L2: SessionStore]
    E --> F{EmotionTrigger?}
    F -->|high emotion| G[L3: EpisodicMemory]
    F -->|normal| H[L4: CoreMemory]

    B --> I[RAG Engine]
    I --> J[FTS5 Search]
    I --> K[MIB Binary Search]
    I --> L[Hybrid RRF Ranking]

    B --> M[Wiki System]
    M --> N[.md Files]
    M --> O[SQLite Index]

    B --> P[Knowledge Graphs]
    P --> Q[Epistemic Graph]
    P --> R[Temporal Graph]

    B --> S[Project Store]
    S --> T[Decisions / Artifacts / Code Index]

    U[Hourly Sweep] -->|consolidate| G
    U -->|promote| H
    U -->|auto-VACUUM| V[(SQLite)]

Comparison

Roadmap

• 4-layer memory hierarchy with layer isolation
• Hybrid search (FTS5 + MIB binary embeddings)
• Knowledge graphs (epistemic + temporal)
• Hourly consolidation sweep + DB self-maintenance
• mcp 2.x native SDK
• Repo rename to Cipher208/a-memory + PyPI package (pip install a-memory)
• Screenshot / asciinema demo in README
• LLM-assisted consolidation on top of the deterministic sweep

Contributing

PRs welcome! See CONTRIBUTING.md.

License

MIT © Cipher208

⭐ If this project helps you, star it on GitHub.