⚡ dakera-mcp

MCP server for Dakera AI. 83 tools. Gives any MCP-compatible AI agent persistent, queryable memory in minutes.

Works with Claude, Claude Code, and any MCP-compatible framework.

Part of Dakera AI — the memory engine for AI agents.

The Dakera memory engine scores 87.8% on LoCoMo (1,540 questions, standard eval) — benchmark details

Install

cargo install dakera-mcp

Or with Docker:

docker pull ghcr.io/dakera-ai/dakera-mcp:latest

Connect

Add to .mcp.json (Claude Code) or claude_desktop_config.json (Claude Desktop):

{
  "mcpServers": {
    "dakera": {
      "command": "dakera-mcp",
      "env": {
        "DAKERA_API_URL": "http://localhost:3300",
        "DAKERA_API_KEY": "your-key"
      }
    }
  }
}

What You Get

83 tools across 15 categories:

• Memory — store, recall, search, decay, importance scoring

• Sessions — create and manage agent sessions

• Agents — namespaces, stats, memory health

• Knowledge — graph construction, entity extraction, clustering, cross-agent network

• Vectors — upsert, query, hybrid search, batch operations

• Full-Text — BM25 index, search, stats

• Operations — health, metrics, backup, audit

Why This Exists

AI agents forget everything when the session ends. Dakera fixes that. This MCP server gives your agent a persistent memory layer with zero infrastructure overhead — point it at a Dakera instance and it works.

→ dakera.ai for hosted instance
→ Self-host with dakera-deploy

Documentation

→ Full docs
→ MCP reference

Related

Part of the Dakera AI open core. The engine is proprietary. The tools are yours.