memorix-memory

memorix

Persistent Memory for AI Agents.
Your agents forget everything between sessions. memorix fixes that.

---

🚀 Quick Start

Want the fastest path? See the TiDB Serverless Quickstart — deploy with 4 environment variables, no separate embedding service needed, estimated 15 minutes to a fully working setup.

Deploying to Kubernetes? See the K8s Deployment Guide — includes Deployment, Service, Ingress, HPA, and ready-to-use manifests in deploy/k8s/.

Server-based memory via memorix-server.

# 1. Deploy memorix-server (Go 1.22+ required)
cd server && MNEMO_DSN="user:pass@tcp(host:4000)/memorix?parseTime=true" \
  MNEMO_LLM_API_KEY="sk-..." \
  MNEMO_LLM_BASE_URL="https://api.openai.com/v1" \
  go run ./cmd/memorix-server

2. Install plugin for your agent (pick one):

Platform	Install
Claude Code	/plugin marketplace add devioslang/memorix then /plugin install memorix-memory@memorix
OpenCode	Add "plugin": ["memorix-opencode"] to opencode.json
OpenClaw	Add memorix to openclaw.json plugins (see openclaw-plugin/README)

# 3. Provision a tenant and set credentials
curl -s -X POST localhost:8080/v1alpha1/memorix
# → {"id":"...", "claim_url":"..."}

export MNEMO_API_URL="http://localhost:8080"
export MNEMO_TENANT_ID="..."

All agents pointing at the same tenant ID share one memory pool.

---

The Problem

AI coding agents — Claude Code, OpenCode, OpenClaw, and others — often maintain separate local memory files. The result:

• 🧠 Amnesia — Agent forgets everything when a session ends
• 🏝️ Silos — One agent can't access what another learned yesterday
• 📁 Local files — Memory is tied to a single machine, lost when you switch devices
• 🚫 No team sharing — Your teammate's agent can't benefit from your agent's discoveries

memorix gives every agent a shared, cloud-persistent memory with hybrid vector + keyword search — powered by TiDB Starter.

Why TiDB Starter?

memorix uses TiDB Starter (formerly TiDB Serverless) as the backing store for memorix-server:

Feature	What it means for you
Free tier	25 GiB storage, 250M Request Units/month — enough for most individual and small team use
TiDB Cloud Zero	Instant database provisioning via API — no signup required for first 30 days
Native VECTOR type	Hybrid search (vector + keyword) without a separate vector database
Auto-embedding (EMBED_TEXT)	TiDB generates embeddings server-side — no OpenAI key needed for semantic search
Zero ops	No servers to manage, no scaling to worry about, automatic backups
MySQL compatible	Migrate to self-hosted TiDB or MySQL anytime

This architecture keeps agent plugins stateless — all state lives in memorix-server, backed by TiDB.

Supported Agents

memorix provides native plugins for major AI coding agent platforms:

Platform	Plugin	How It Works	Install Guide
Claude Code	Hooks + Skills	Auto-loads memories on session start, auto-saves on stop	claude-plugin/README.md
OpenCode	Plugin SDK	system.transform injects memories, session.idle auto-captures	opencode-plugin/README.md
OpenClaw	Memory Plugin	Replaces built-in memory slot (kind: "memory"), framework manages lifecycle	openclaw-plugin/README.md
Any HTTP client	REST API	curl to memorix-server	API Reference