brain-server

markfietje/brain-server

★ 0 stars Rust AI/LLM Updated 3d ago

Local-first semantic-memory and knowledge-graph server for AI agents. Rust + hybrid RAG retrieval, temporal facts, human-in-the-loop governance, append-only audit chain. Self-hosted, offline-first, zero per-query cost, no LLM in the loop. OpenAI-compatible API, MCP server.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "brain-server": {
      "command": "cargo",
      "args": [
        "run",
        "--",
        "brain-server"
      ]
    }
  }
}

README Excerpt

**A local-first semantic memory and knowledge-graph server for AI agents. No cloud, no per-query cost, no LLM in the loop.** Brain Server gives your agent a second brain that lives on your own device. It is written in Rust and wraps a deterministic retrieval engine, a static local embedding model, and a knowledge graph behind a versioned HTTP API. Recall never calls an LLM or an embedding API, so every query costs nothing and the data never leaves the machine.

Tools (6)

BRAIN_AUDIT_READ_EVENTSBRAIN_AUDIT_RETENTION_DAYSBRAIN_DB_PATHBRAIN_JWT_ISSUERBRAIN_SUGGEST_ENABLEDINJECTION_POLICY

Topics

ai-agentsarmedge-aiembeddingsknowledge-graphlocal-firstmemoryoffline-firstragrustsemantic-searchvector-search

brain-server

Quick Install

README Excerpt

Tools (6)

Topics

Related AI/LLM Servers