Skip to main content

MCP server for AgentRecall — persistent memory for AI agents

Project description

AgentRecall MCP Server

MCP server that gives your AI agents persistent memory. Works with Claude Code, Claude Desktop, Hermes Agent, OpenClaw, and any MCP-compatible client.

Quick Start

# Install
pip install agentrecall-mcp

# Run (cloud mode)
AGENTRECALL_API_KEY=ark_your_key agentrecall-mcp

Setup by Client

Claude Code

claude mcp add --transport stdio --scope user \
  --env AGENTRECALL_API_KEY=ark_YOUR_KEY_HERE \
  agentrecall -- agentrecall-mcp

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "agentrecall": {
      "command": "agentrecall-mcp",
      "env": {
        "AGENTRECALL_API_KEY": "ark_YOUR_KEY_HERE"
      }
    }
  }
}

Restart Claude Desktop after saving.

OpenClaw

openclaw mcp set agentrecall '{"command":"agentrecall-mcp","env":{"AGENTRECALL_API_KEY":"ark_YOUR_KEY_HERE"}}'

Hermes Agent

Add to ~/.hermes/config.yaml:

mcp_servers:
  agentrecall:
    command: "agentrecall-mcp"
    env:
      AGENTRECALL_API_KEY: "ark_YOUR_KEY_HERE"

Then restart: hermes gateway restart

Any MCP Client

The server uses stdio transport. Point your MCP client to:

AGENTRECALL_API_KEY=ark_YOUR_KEY agentrecall-mcp

Configuration

Environment Variable Default Description
AGENTRECALL_API_KEY Your AgentRecall API key (required, agent-specific)
AGENTRECALL_API_URL https://api.agentrecall.cloud/v1 API endpoint

How Agent-Specific Keys Work

API keys are bound to specific agents. When you create a key in the dashboard, you select which agent it belongs to. This means:

  • No agent_id needed — The key remembers which agent it's for
  • Better security — Each key only accesses one agent's memories
  • Multiple keys — Create separate keys for different environments (dev, staging, prod)

Available Tools

Tool Description
store_memory Store a memory (auto-classified by AI)
search_memories Semantic search across memories
get_memory Get a memory by ID
update_memory Update memory content
delete_memory Delete a memory
list_memories List memories for an agent
count_memories Count memories
traverse_graph Traverse the knowledge graph
get_graph_stats Get graph statistics
find_graph_path Find path between two entities
get_graph_context Get graph context for a query
list_agents List all agents
create_agent Create a new agent

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agentrecall_mcp-2.1.0-py3-none-any.whl (6.2 kB view details)

Uploaded Python 3

File details

Details for the file agentrecall_mcp-2.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for agentrecall_mcp-2.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 157df5c9ad14b99adfe3958326837b23b9fcefc2b0e2ae088e8cfa8cd97520f7
MD5 1256d5158b5a8fb52747ca9b216175d1
BLAKE2b-256 471dd5ad5209849160a198a56c3b57d74049614024dc71bd0e05fdd40bfbe06a

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page