@agenticstore
AgenticStore: The secure toolkit for AI agents. Instantly equip Claude Desktop, Cursor, and Windsurf with 27+ MCP tools, persistent memory, and SearXNG search, all protected by a built-in PII prompt firewall to protect your data from being exposed to AI agents.
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"agenticstore": {
"command": "uvx",
"args": [
"agentic-store-mcp"
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。
点击上方按钮,在临时沙箱中运行此服务器并实时获取工具列表。
Agenticstore is an open-source Model Context Protocol (MCP) server that provides an LLM prompt firewall, token optimization, and AI security toolkit. It offers 31 agent tools for data privacy, prompt recording, audit logs, and AI assistant integration with Claude Code, Cursor, and Windsurf.
Install via pip install agentic-store-mcp --upgrade or uvx agentic-store-mcp, then run agentic-store-mcp. Configure your AI client manually in its config file or use the bundled GUI webapp (agentic-store-webapp at http://localhost:8765) for visual management. Quick start options include V0 (PyPI), V1 (GitHub source), and V2 (MCP Hub UI).
Python 3.12+ is required. The firewall feature is currently tested and stable on macOS; Linux and Windows support is coming soon. Optional integrations include Ollama for local LLM sanitization and SearXNG for web search.
Yes, it is fully free with no accounts, paywalls, or subscriptions. Licensed under MIT.
It provides 31 MCP tools, divided into 22 local-only tools (no external API), 7 tools requiring remote APIs, 6 with write access, and 25 read-only tools.
No. Every operation runs locally on your machine. The prompt firewall intercepts and audits traffic locally before it reaches any external API.
It supports Claude Code (CLI), Cursor, and Windsurf. Configuration can be done manually in client config files or automatically via the GUI webapp.
An MCP server that allows you to run TypeScript, JavaScript, and Python code in a sandbox on your local machine using the Deno® sandbox. This server provides a controlled environment for executing code with explicit permission controls.
Python sandboxes for llms
An MCP server allowing AI assistants to use a Neo4j knowledge graph as their primary, dynamic instruction manual and long term project memory with adaptive templating and autonomous tool development tools.
MCP server that allow LLM in agent mode builds n8n workflows for you
Turn any JavaScript module into a sandboxed MCP (Model Context Protocol) server with automatic reflection and type inference.
评论