에이전트 강화
게시자의 자체 명령입니다. PlugMCP 엔드포인트가 아닙니다.
Host a copy with our keys and logs.
mcpb https://github.com/mahaoboy/memalyze/releases/download/v0.2.0/memalyze-0.2.0-darwin-arm64.mcpbClaude Code에는 로컬 명령용 설치 링크가 없습니다. 아래 명령을 복사해 터미널에서 실행합니다.
claude mcp add io.github.mahaoboy-memalyze -- mcpb https://github.com/mahaoboy/memalyze/releases/download/v0.2.0/memalyze-0.2.0-darwin-arm64.mcpb터미널에서 실행합니다.
{ "mcpServers": { "io.github.mahaoboy-memalyze": { "command": "mcpb", "args": [ "https://github.com/mahaoboy/memalyze/releases/download/v0.2.0/memalyze-0.2.0-darwin-arm64.mcpb" ] } }}claude_desktop_config.json
원격 URL인가요, 내 컴퓨터에서 실행하는 명령인가요? 두 설치 방식의 차이
PlugMCP에서 실행
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
실행게시자 제공
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 · ·
Self-hosted agent memory: redaction before write, append-only audit, local embeddings.
./memalyze # MCP over stdio, for Claude Code / Cline / any harness
v0.1.0 is out — static binaries for linux/darwin/windows, checksums published alongside.
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:
memalyze is the middleware that fixes those three, and nothing else.
* 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.
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:
[REDACTED:phone], not 13800138000);Details and transcripts: docs/s15-redaction-audit.md.
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.
Claude Code (project scope):
claude mcp add --scope project memalyze -- "$(pwd)/memalyze"
cp examples/claude-code/CLAUDE.md ./CLAUDE.md # makes Claude search before answering
Cline: copy examples/cline/cline_mcp_settings.json and .clinerules.
Generic harness: examples/custom-harness/ — one prompt contract
plus pseudocode. All options: examples/README.md.
Download the bundle for your platform from the
v0.1.0 release and open it.
Claude Desktop shows an install sheet and asks where to keep memalyze.db.
The binary inside is the same v0.1.0 build as above. These bundles are new and have not
been opened on a real macOS/Windows desktop yet — if the install sheet does not appear,
use the plain binary. The server is also listed in the
MCP Registry.
printf 'MEMALYZE_HTTP_TOKEN=%s\n' "$(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.
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).
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.
Modes: enforce (default), dry-run (count only, for tuning), off (debugging).
Custom rule packs are JSON: examples/redaction-rules.json.
Audit (read path). Append-only audit_log (a database trigger rejects UPDATE):
actor, tenant, scope, action, memory id, timestamp, result count, degraded, redaction
count. The search text itself is stored as SHA-256 + length unless you explicitly set
MEMALYZE_AUDIT_QUERY_PLAIN=1.
memalyze audit export --from 2026-10-01 --to 2026-10-31 --format csv --out audit.csv
memalyze audit export --tenant acme --format json
What is not here yet: per-user authentication (one token per process, actor from
env), audit retention/purge, reversible redaction, audit UI.
Numbers from S8 on a 60-memory Chinese corpus, 16 queries, measured not assumed
(docs/s8-retrieval-report.md):
Two counter-intuitive findings that shaped the design: SQLite FTS5's default tokenizer cannot search Chinese at all (0.19 recall until we stored bigram text), and RRF fusion scored worse than vector-only, so v0.1 does not fuse. See D-012.
Latency (S14, cold start fixed): warmup 37.8 s in background, then search 1.14 s with
degraded=false; after five idle minutes a fresh process answers in 1.52 s. If the vector
path is unavailable, search degrades to keyword in ≤3 s instead of hanging (D-015/D-018).
harness (Claude Code / Cline / your agent)
│ MCP: stdio or Streamable HTTP
▼
internal/mcp 4 tools, argument conversion only
▼
internal/memory redact → dedup → store → embed; search: vector → keyword → degraded
├── internal/govern rule engine (8 families, JSON configurable)
├── internal/store interface; sqlite impl (memories, embeddings, audit_log)
├── internal/embed float32 BLOB codec, cosine; Ollama client
└── internal/search CJK bigram tokenizer + FTS5 MATCH builder
float32 little-endian BLOBs; search is brute-force cosine (fine under ~10k
memories, and the honest limit of a single-binary design).PRAGMA user_version (v1 → v2 adds the audit table).memalyze audit export --from 2026-10-01 --format csv --out audit.csv # who read what
memalyze backup --out /var/backups/memalyze-$(date +%F).db # consistent snapshot
memalyze health --url http://127.0.0.1:8080/healthz # probe
deploy/systemd/memalyze.service +
docs/s16-systemd.md (hardened unit, env file, backup timer).docs/s16-docker.md.backup uses VACUUM INTO, so snapshots are consistent under WAL and include the audit
table; it refuses to overwrite an existing file.All through environment variables; defaults work out of the box.
Honest snapshot — this is an early-stage project, not a 1.0.
Known gaps: no per-user auth, no audit retention, the Docker image has not been built in CI (no daemon in the dev environment), brute-force vector scan, Chinese-first defaults (English works, but the tokenizer is tuned for CJK).
Two-week pilot: we install memalyze in your environment, tune the redaction rules to your data, review the audit export with you, and write down the threat-model gaps we find. No fee; you keep the deployment and the notes.
AGPL-3.0-or-later (LICENSE). Commercial licensing for embedding in closed products:
open an issue titled commercial license.
Contributions: see CONTRIBUTING.md. Security reports: see
SECURITY.md.
Self-hosted agent memory: redaction before write, append-only audit, local embeddings.
{ "mcpServers": { "io.github.mahaoboy-memalyze": { "command": "mcpb", "args": [ "https://github.com/mahaoboy/memalyze/releases/download/v0.2.0/memalyze-0.2.0-darwin-arm64.mcpb" ] } }}도구
아직 도구 목록이 없습니다. 원격 서버는 첫 상태 확인 후에 도구 목록이 생깁니다.
FAQ
이 페이지에서 Cursor, Claude Desktop, Claude Code 또는 VS Code용 stdio 설정을 복사합니다. PlugMCP 프록시가 아니라 게시자의 설치 방식을 사용합니다.
이 레코드에는 필수 환경 변수가 등록되어 있지 않습니다.
Memalyze은(는) stdio를 통해 로컬에서 실행됩니다.
마지막 확인: 2026-11-15. Self-hosted agent memory: redaction before write, append-only audit, local embeddings.