Mneme

Persistent Memory for AI Agents — MCP-Native. Local-First. Zero Dependencies.

Mneme is a single Rust binary that gives AI agents durable memory across sessions.
One binary. One file. No Docker. No Postgres. No cloud. Just persistent memory
that works with any MCP host.

One-Line Install

curl -sSf https://raw.githubusercontent.com/Perseus-Computing-LLC/mneme/main/scripts/install.sh | sh

That's it. Mneme is installed to ~/.local/bin/mneme. Start it:

mneme serve --db ~/.mimir/data/mneme.db

macOS note. On Apple Silicon, an unsigned binary is killed on launch
(Killed: 9, no output) by the OS binary policy — even with no quarantine
attribute. The installer ad-hoc code-signs Mneme for you. If you build or copy
the binary yourself (cargo build --release && cp target/release/mneme ~/.cargo/bin/), sign it once after each rebuild:

codesign --sign - "$(command -v mneme)"

Connect any MCP host (Claude Desktop, Cursor, Hermes Agent, Perseus, etc.):

{
  "mcpServers": {
    "mneme": {
      "command": "mneme",
      "args": ["serve", "--db", "~/.mimir/data/mneme.db"]
    }
  }
}

30-Second Quickstart

# Start Mneme
mneme serve --db memory.db &
sleep 1

# Remember a fact (via MCP JSON-RPC on stdio)
echo '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"mimir_remember","arguments":{"category":"demo","key":"hello","body_json":"{\"text\":\"Hello from Mneme!\"}"}}}' | mneme serve --db memory.db

# Search for it
echo '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"mimir_recall","arguments":{"query":"Hello"}}}' | mneme serve --db memory.db

Works With Every MCP Client

Mneme is a standard MCP stdio server — the same mneme serve command works
everywhere. Run mneme doctor to validate your install and print this matrix locally.

Copy-paste config snippets for each: docs/clients/.

Why Mneme

Mneme is the only memory engine that is simultaneously MCP-native,
local-first, zero-dependency, AND agent-first.

Comparison Matrix

Full comparison: Mneme vs Mem0 →
vs Letta →
vs Zep →

Stress Test: 100K Entities

Mneme handles production workloads on modest hardware:

Run it yourself: cargo test stress_100k --release -- --ignored --nocapture

Framework Integrations

Ready-to-use adapters that make Mneme the default memory backend for
popular AI agent frameworks:

Each adapter:

• Connects via MCP stdio subprocess (persistent session)
• Maps the framework's memory interface to Mneme tools
• Comes with a README quickstart (5 minutes to working)
• Has passing tests with mocked MCP transport

Any MCP-compatible framework works with Mneme directly. See
Awesome Mimir for the full list.

46 MCP Tools

Entity CRUD

Search & RAG

Graph

Journal

State

Lifecycle

Quality

Vault & Federation

Metrics & Ops

CLI

# Server
mneme serve --db /data/mneme.db
mneme serve --web --port 8767 --encryption-key ~/.mimir/secret.key
mneme serve --llm-endpoint http://localhost:11434/api/generate --llm-model llama3
mneme serve --transport sse --port 8787 --mcp-token my-secret-token

# Maintenance (operate directly on DB, no server needed)
mneme stats          --db /data/mneme.db
mneme forget         --db /data/mneme.db --category decision --key stale-choice --reason "superseded"
mneme prune          --db /data/mneme.db --category junk --min-decay 0.1 --dry-run
mneme purge          --db /data/mneme.db --dry-run
mneme decay          --db /data/mneme.db
mneme reindex        --db /data/mneme.db
mneme vault-export   --db /data/mneme.db --vault-dir ./export/
mneme vault-import   --db /data/mneme.db --vault-dir ./export/
mneme obsidian-sync  ~/obsidian-vault/Mneme/          # one-shot export to an Obsidian vault
mneme obsidian-sync  ~/obsidian-vault/Mneme/ --watch  # continuous sync on every memory change

# Key management
mneme keygen --key-file ~/.mimir/secret.key

Flags

Your AI Memory in Obsidian

Mneme is your AI agent's long-term memory — and it doubles as your second
brain. Every entity your agent remembers exports to a plain Markdown note with
YAML frontmatter, so your AI's memory becomes a navigable personal knowledge
base inside the tools you already use: Obsidian, Logseq, or Notion.

# Export your entire memory to an Obsidian vault as linked Markdown notes
mneme obsidian-sync ~/obsidian-vault/Mneme/

# Keep it live — re-export automatically on every memory change
mneme obsidian-sync ~/obsidian-vault/Mneme/ --watch

Open the vault in Obsidian and you get a graph of your agent's knowledge.

WikiLink backlinks. When one entity links to another (via mimir_link or a
depends_on / implements / references relationship), the exported note gets
a ## Links section with [[WikiLink]] backlinks that resolve natively in
Obsidian's graph view:

---
id: cli-de8dfb8364b6
category: architecture
key: api
type: insight
decay_score: 0.5000
---

{"content":"axum service"}

## Links

- [[cli-99756b494c7d|database]] (depends_on)

Links resolve by entity id (notes are written as <id>.md) so they never
break, and Obsidian shows the human-readable key as the link label. Open the
graph view and your agent's architecture, decisions, and insights become a
clickable knowledge map.

--watch polls Mneme's cheap, deterministic state digest on an interval and
re-exports only when memory actually changes. It naturally catches every
mimir_remember write with no filesystem-watcher dependency and no coupling to
the server. Tune the interval with MIMIR_SYNC_INTERVAL_SECS (default: 2s).

Other PKM tools

Unlike cloud-only "second brain" tools, Mneme runs 100% local, is written in
Rust, encrypts at rest with AES-256-GCM, and applies decay scoring so
stale memories fade — your knowledge base stays yours and stays fresh.

Features

Semantic Search (on by default)

• Bundled, in-process embeddings — a quantized all-MiniLM-L6-v2 model
  (384-dim) is compiled into the binary, so dense/semantic search works with
  zero config and zero network: no Ollama, no API key, no model download.
  This is the default build (bundled-embeddings feature).
• Auto-embed on write (#271) — mimir_remember embeds each new (or
  content-changed) entity synchronously as it is written, using the bundled
  model. Single-entity embedding is deterministic and LRU-cached, so it is cheap
  and adds no background tasks. Embedding failures are non-fatal (logged to
  stderr); the write always succeeds.
• Hybrid is the default recall mode (#271) — mimir_recall(query=...) with
  no mode flag automatically selects hybrid (dense + keyword fused via RRF)
  whenever embeddings exist, and transparently falls back to fts5 keyword
  search when none do. No manual mimir_embed step, no flags to remember.
• mimir_semantic_search(query, limit) — a one-tool shortcut for pure
  dense, meaning-based search (no keyword fallback) when you just want "find
  things like this".
• Optional alternate embedder — to use Ollama or any OpenAI-compatible
  /v1/embeddings endpoint instead of the bundled model, set --llm-endpoint
  (and --embedding-endpoint / --llm-api-key as needed). This is entirely
  optional; the bundled model is used by default.
• Build a lean binary without bundled embeddings via
  cargo build --no-default-features — recall then defaults to keyword search
  unless a remote embedder is configured.

Hybrid Search internals

• FTS5 keyword search with LIKE fallback and Porter stemming expansion
• Dense vector search via cosine similarity on stored embeddings
• Reciprocal Rank Fusion (RRF) — combine keyword + vector results
• Query expansion — automatic stemming variants for broader recall

Memory Lifecycle

Mneme models memory using three biomimetic layers, inspired by human memory pathways:

• World (Core): Slow-decaying, global facts about the environment.
• Episodic (Buffer): Fast-decaying, session-specific interaction history.
• Semantic (Working): Medium-decaying, general knowledge and learned concepts.

You can interact with these layers directly using the mimir_recall_layer tool or by specifying the layer parameter in mimir_remember.

• Ebbinghaus decay — memories naturally fade unless retrieved (refresh on access)
• Layer promotion — buffer → working → core based on access frequency
• Automatic archival — stale entities archive; purge to permanently delete + VACUUM
• Always-on entities — pin critical memories for unconditional session injection

RAG & Embeddings

• mimir_ask — natural language Q&A over stored memories via any LLM (Ollama, OpenAI, etc.)
• mimir_embed — generate and store dense vectors via Ollama or OpenAI-compatible /v1/embeddings
• Supports single-entity and batch-category embedding

Encryption

• AES-256-GCM transparent encryption for entity body_json
• Opt-in via --encryption-key flag
• mneme keygen subcommand for key generation
• FTS5 index stays plaintext for search

Web Dashboard

• Built-in Axum HTTP server (mneme serve --web --port 8767)
• Dark-themed dashboard with search, entity table, vis.js graph, timeline
• Default bind: 127.0.0.1 (use --web-bind 0.0.0.0 to expose)
• Separate SQLite connection in WAL mode for concurrent reads

External Connectors

• GitHub issues connector — ingest issues/PRs by repo, rate-limit aware
• File watcher — scan directories for .md/.txt/.json files with content-hash dedup
• YAML-based connector config via --connectors-config

Multi-Transport

• stdio (default) — zero-config, works with any MCP host
• SSE — Server-Sent Events for HTTP-based MCP clients
• HTTP — REST-style MCP endpoint
• Bearer token auth — for SSE/HTTP transports

Perseus Integration

Mneme is the default memory backend for Perseus:

mimir:
  enabled: true
  transport: "stdio"
  command: ["mneme", "serve", "--db", "~/.mimir/data/mneme.db"]
  timeout_s: 30.0
  merge_strategy: "local_first"
  fallback_to_local: true
  context_categories: ["decision", "architecture", "convention"]
  context_limit: 10

Government & Federal Procurement

Mneme is built for government deployment from the ground up.

For federal buyers: See docs/federal-buyers.md for
procurement information, compliance status, and deployment models (air-gapped,
on-premises, classified environments).

Perseus Computing LLC is a US-owned small business. SAM.gov registration in progress.
NAICS: 541715, 541511, 541512.

Privacy Policy

Mneme is a local-first MCP server — it runs entirely on your machine.

Data Collection

• No data collection. Mneme does not collect, transmit, or phone home any user data, usage statistics, or telemetry.
• All data remains in your local SQLite database file.

Data Usage & Storage

• All memory entities, journal entries, and state are stored locally in a SQLite database at the path you specify via --db.
• Optional AES-256-GCM encryption at rest is available — when enabled, entity bodies are encrypted before storage.
• No data is shared with Perseus Computing LLC or any third party.

Third-Party Sharing

• None. Mneme is fully air-gapped by default. No API calls, no cloud services, no external network requests.
• The optional dense vector embeddings feature uses a locally-compiled model — no external embedding API is called.

Data Retention

• You control retention: entities can be soft-deleted (mimir_forget), archived (via decay/compact), or permanently purged (mimir_purge).
• No automatic off-machine backup is performed.

Contact

• Email: [email protected]
• GitHub: Perseus-Computing-LLC/mneme

License

MIT — see LICENSE.