Agentdrive
@Rakesh1002
Agentdrive について
Official MCP server for AgentDrive — cross-agent artifact workspace with provenance. 20 tools over stdio or hosted at mcp.agentdrive.sh.
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"agentdrive": {
"command": "npx",
"args": [
"-y",
"@adrive/mcp-server"
],
"env": {
"AGENTDRIVE_API_KEY": "<YOUR_AGENTDRIVE_API_KEY>"
}
}
}
}ツール
20memory
memory
memory
files
files
files
files
files
collaboration
collaboration
collaboration
provenance
provenance
org
org
org
sandbox (experimental)
sandbox (experimental)
sandbox (experimental)
sandbox (experimental)
概要
What is Agentdrive?
Agentdrive is the official MCP server for AgentDrive, a cross-agent artifact workspace with provenance and portable storage. It integrates with Claude Code, Codex, Cursor, LangGraph, and any MCP-aware host to give agents a shared, persistent workspace with versioning, semantic recall, run logging, and cross-agent attribution.
How to use Agentdrive?
You can connect via a hosted Streamable HTTP endpoint (https://mcp.agentdrive.sh/mcp) using a bearer token, or run locally via npx with the @adrive/mcp-server package and the AGENTDRIVE_API_KEY environment variable. The fastest setup is the AgentDrive installer (curl -fsSL agentdrive.sh/install | sh), which auto-configures detected MCP hosts.
Key features of Agentdrive
- Auto-versioned file storage with read/write/delete/list
- Semantic search and recall across workspace artifacts
- Cross-agent provenance and run logging
- Human collaboration tools (share links, comments, feedback)
- Org management (invite, roles, view details)
- Experimental Linux sandbox for git clone/exec/commit
Use cases of Agentdrive
- Hand off work between Claude Code and Cursor with full context preserved
- Recall previous agent outputs and decisions across sessions
- Share generated artifacts with teammates via gated links
- Log and attribute agent runs for auditing or debugging
FAQ from Agentdrive
What is Agentdrive?
Agentdrive is an MCP server that provides a cross-agent artifact workspace. It gives agents a shared persistent workspace with versioned files, semantic recall, and provenance across different AI coding tools.
How do I get an API key?
Create one for free at https://app.agentdrive.sh. No credit card is required. The free tier includes 1 workspace, 1 GB storage, and 10k embeddings
「AI とエージェント」の他のコンテンツ
1MCP - One MCP Server for All
1mcp-appA unified Model Context Protocol server implementation that aggregates multiple MCP servers into one.
MCP-LLM Bridge
patruffBridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools
1Panel
1Panel-dev🔥 1Panel is a modern, open-source VPS control panel — and the only one with native AI agent support. Run Ollama models, deploy OpenClaw agents, and manage your entire server stack from one clean web interface.
MCP Manager for Claude Desktop
zueaisimple web ui to manage mcp (model context protocol) servers in the claude app
MCP Server - Remote MacOs Use
baryhuangThe only general AI agent that does NOT requires extra API key, giving you full control on your local and remote MacOs from Claude Desktop App
コメント