CLAUDE-MEM GOES SERVER-SIDE (AND APACHE-2.0): OPEN-SOURCE AGENT MEMORY GROWS UP
claude-mem 13 adds an opt-in server runtime with Postgres/Redis and a /v1 API, and switches from AGPL to Apache-2.0. The v13 release lands a Postgres-backed ob...
claude-mem 13 adds an opt-in server runtime with Postgres/Redis and a /v1 API, and switches from AGPL to Apache-2.0.
The v13 release lands a Postgres-backed observation store, optional BullMQ/Redis queues, API-key auth, an outbox pattern, Docker Compose, and a new /v1 REST surface — all behind an opt‑in Server Beta, plus a relicense to Apache-2.0 for easier adoption release notes.
At the same time, enterprise ops patterns are clearly forming: Uber describes running 60k agent tasks/week on MCP with service-like controls and auditability talk. Tooling is catching up too — MassGen added a TUI consensus map for multi-agent coordination v0.1.84 and Antigravity’s skills pack shipped production-oriented agent workflows, including on-chain transaction orchestration v11.0.0.
Relicensing to Apache-2.0 removes a major legal blocker for enterprise use and vendor embedding.
A server runtime with Postgres/Redis and a REST API makes agent memory observable, auditable, and ops-friendly.
-
terminal
Spin up the Server Beta via Docker Compose, write events through /v1, and verify outbox-to-queue delivery plus idempotent job execution.
-
terminal
Benchmark memory search latency and queue throughput on your hardware; compare worker-era SQLite vs Postgres+Redis under load.
Legacy codebase integration strategies...
- 01.
Pilot the server alongside the existing worker plugin and mirror writes; validate auth, RBAC, and data residency/backups in Postgres.
- 02.
Scope a phased migration by mapping current memory/events to the new /v1 models; test multi-tenant API keys and audit export.
Fresh architecture paradigms...
- 01.
Design agents as services: standardize on /v1 for events/sessions and wire MCP tools to a central memory plane.
- 02.
Use Postgres schemas for observability (sessions, context, audit) and Redis queues for backpressure and retries.
Get daily POSTGRESQL + SDLC updates.
- Practical tactics you can ship tomorrow
- Tooling, workflows, and architecture notes
- One short email each weekday