memalyze

memalyze

MCP-native, self-hosted memory for AI agents — with the governance an enterprise will actually ask for.

One static binary. Local embeddings. No cloud, no graph database, no Python stack. Redaction happens before anything is written; every read is audited.

Website · MCP Registry · v0.1.0

./memalyze          # MCP over stdio, for Claude Code / Cline / any harness

v0.1.0 is out — static binaries for linux/darwin/windows, checksums published alongside.

---

The problem

Every agent framework ships its own memory: a vector table here, a JSON file there, a "notes" folder somewhere else. Three things follow, and all three hurt in production:

1. You cannot audit it. "Which customer's data did the agent read last Tuesday?" has no answer, so security review stops the project.
2. You cannot deploy it offline. Most options either call a cloud API or download model weights from HuggingFace at runtime — both fail inside a bank, a hospital, or a factory.
3. You cannot keep it small. Twenty-three tools, a graph database, and a Python dependency tree later, the "memory feature" is a second product to operate.

memalyze is the middleware that fixes those three, and nothing else.

What is different

\* A database trigger rejects UPDATE on audit_log; DELETE is not blocked, so anyone with the database file can remove rows. It is append-only at the SQL level, not WORM — export to object-lock or syslog storage if deletions are inside your threat model (SECURITY.md).

Not another mem0, and not Basic Memory either: those are good at *remembering*. memalyze assumes remembering is becoming a commodity and competes on governance, offline operation, and a deliberately small surface.

30-second demo

No Ollama, no network, no MCP client:

./examples/demo/demo.sh

or: go run ./cmd/memalyze demo

It writes a memory containing a phone number, an e-mail and a resident ID, then shows:

• what is actually stored ([REDACTED:phone], not 13800138000);
• that searching for the original number returns 0 hits;
• that a business-word search still hits, returning the redacted text;
• the audit rows for the two writes and the two reads.

Details and transcripts: docs/s15-redaction-audit.md.

Quick start

1. Binary + local embedding model

Prebuilt, checksum-verified (Linux/macOS; Windows: take the .exe from the release page):

curl -fsSL https://raw.githubusercontent.com/mahaoboy/memalyze/main/scripts/install.sh | sh

ollama pull qwen3-embedding:0.6b # ~640 MB, 1024-dim, strong on Chinese memalyze --version

From source (Go 1.27+): make build, then ./memalyze --version.

2. Wire it into a harness

Claude Code (project scope):

claude mcp add --scope project memalyze -- "$(pwd)/memalyze"

cp examples/claude-code/CLAUDE.md ./CLAUDE.md # makes Claude search before answering

Cline: copy examples/cline/cline_mcp_settings.json and .clinerules. Generic harness: examples/custom-harness/ — one prompt contract plus pseudocode. All options: examples/README.md.

3. Or one-click in Claude Desktop (MCPB)

Download the bundle for your platform from the v0.1.0 release and open it. Claude Desktop shows an install sheet and asks where to keep memalyze.db.

The binary inside is the same v0.1.0 build as above. These bundles are new and have not been opened on a real macOS/Windows desktop yet — if the install sheet does not appear, use the plain binary. The server is also listed in the MCP Registry.

4. Or run it in Docker (memalyze + Ollama + model pull)

printf 'MEMALYZE_HTTP_TOKEN=%s\n' "$(openssl rand -hex 24)" > .env

docker compose up -d --build curl -s http://127.0.0.1:8080/healthz # {"status":"ok","version":"dev"}

Point a Streamable-HTTP MCP client at http://127.0.0.1:8080/mcp with Authorization: Bearer $MEMALYZE_HTTP_TOKEN. See docs/s16-docker.md.

Tools

Four, on purpose. Every extra tool is a decision the model has to make.

There is no memory_context tool: memory_search already returns text you can paste into a prompt, and two overlapping tools make model choice worse (D-017).

Governance

Redaction (write path). Rules run in memory.Service.Add *before* dedup, FTS and embedding — so plaintext never reaches the main table, the full-text index or a vector.

| private_key_block | `[PopNx MCP ตัดค่าที่หน้าตาเหมือนกุญแจออก]