MCP Server

Dakera ships as a native MCP (Model Context Protocol) server, giving any AI assistant persistent memory, semantic search, and knowledge graphs through 14 core tools loaded by default. 86 additional tools are available via profiles when you need them — keeping your agent's context window lean (~2,964 tokens instead of ~16K).

Quick start

Get from zero to working memory in under 5 minutes.

1. Set your environment

Two environment variables connect dakera-mcp to your Dakera server:

export DAKERA_API_URL=http://<YOUR_SERVER_IP>:3300
export DAKERA_API_KEY=<your-api-key>

2. Install dakera-mcp

npm install -g @dakera-ai/dakera-mcp
# or run without installing:
npx @dakera-ai/dakera-mcp

brew install dakera-ai/tap/dakera-mcp

cargo install dakera-mcp       # Rust 1.70+
dakera-mcp --version

docker run --rm -i \
  -e DAKERA_API_URL=$DAKERA_API_URL \
  -e DAKERA_API_KEY=$DAKERA_API_KEY \
  ghcr.io/dakera-ai/dakera-mcp:latest

3. Verify connectivity

dakera-mcp --test-connection
# Connected to Dakera v0.11.83 — 14 core tools loaded (86 available)

4. Connect your AI client

Add Dakera to your client's MCP config. Replace placeholders with your actual values:

// Claude Desktop: ~/Library/Application Support/Claude/claude_desktop_config.json
// Claude Code: .mcp.json (project) or ~/.claude/settings.json (global)
// Cursor: ~/.cursor/mcp.json
// Windsurf: ~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "dakera": {
      "command": "dakera-mcp",
      "env": {
        "DAKERA_API_URL": "http://<YOUR_SERVER_IP>:3300",
        "DAKERA_API_KEY": "<your-api-key>"
      }
    }
  }
}

Restart your client after editing. For Zed, use context_servers with a command.path key instead of mcpServers.

5. Test it

Ask your AI assistant: "Store a memory that I prefer dark mode" — it should call dakera_store and confirm the memory was saved.

Profiles & tool selection

Dakera organizes its 86 tools into four profiles so you only pay the token cost for what you use. The default core profile covers everything most agents need.

Which profile do I need?

Setting your profile

# Option 1: Environment variable (recommended)
DAKERA_MCP_PROFILE=power    # Options: core, power, admin, all

# Option 2: Per-request parameter (in tools/list params)
{"profile": "power"}

# Option 3: Default — core (14 tools, ~2,964 tokens)

Profile overview

On-demand tool discovery

Instead of switching profiles, you can discover and load individual tools at runtime using the two meta-tools (always available in every profile):

# Step 1: Browse available tools by keyword or tier
dakera_discover_tools(tier="power")
# → returns tool names + descriptions (no schemas loaded, no token cost)

# Step 2: Load schemas for the tools you need
dakera_load_tools(tools=["dakera_consolidate", "dakera_agent_stats"])
# → returns full inputSchema for each tool — now callable in this session

# Step 3: Use the tool normally
dakera_consolidate(agent_id="my-agent", memory_ids=["mem_1", "mem_2"])

Core tools by workflow

The 14 core tools organized by what you're trying to do, not by implementation category.

Memory: store, find, and manage

The fundamental memory CRUD operations — store information, find it later, and clean up what you don't need.

Search: hybrid and full-text

Beyond semantic recall, Dakera offers BM25 keyword search and hybrid search that combines both approaches.

Sessions: group related memories

Sessions let you group memories from a single conversation or task, making it easy to recall context from a specific interaction.

Knowledge graphs

Build, traverse, and query relationship graphs between memories. Build graphs from embedding similarity, then traverse or export them.

Meta-tools: discover and load

Always available in every profile. Use these to explore and dynamically load tools from higher tiers.

Full tool reference by tier

All 86+ tools organized by tier and workflow. Core tools are documented in detail above — this section provides a quick reference for all tiers.

Power tier (68 tools)

Everything in core, plus:

Admin tier (32 tools)

Everything in core, plus namespace and policy management:

All tier (86+ tools)

Everything above, plus low-level vector operations for custom embedding workflows: