{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_243u42kgdqyc","handle":"iflow-mcp-tomconte-codebox-ai","url":"https://wellknown.network/agents/iflow-mcp-tomconte-codebox-ai","links":{"self":"https://wellknown.network/agents/iflow-mcp-tomconte-codebox-ai/record.json","html":"https://wellknown.network/agents/iflow-mcp-tomconte-codebox-ai","markdown":"https://wellknown.network/agents/iflow-mcp-tomconte-codebox-ai/record.md","api":"https://wellknown.network/api/v1/agents/iflow-mcp-tomconte-codebox-ai","status":"https://wellknown.network/api/v1/agents/iflow-mcp-tomconte-codebox-ai/status","claim":"https://wellknown.network/agents/iflow-mcp-tomconte-codebox-ai/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/iflow-mcp-tomconte-codebox-ai/claim.json","badge":"https://wellknown.network/agents/iflow-mcp-tomconte-codebox-ai/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/iflow-mcp-tomconte-codebox-ai/history","tools":"https://wellknown.network/api/v1/agents/iflow-mcp-tomconte-codebox-ai/tools"},"ard":{"identifier":"urn:air::server:iflow-mcp-tomconte-codebox-ai","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"iflow-mcp_tomconte-codebox-ai","summary":"A secure Python code execution service that provides a self-hosted alternative to OpenAI's Code Interpreter","description":"# CodeBox-AI\n\nA secure Python code execution service that provides a self-hosted alternative to OpenAI's Code Interpreter or Anthropic's Claude analysis tool. Built with FastAPI and IPython kernels, it supports session-based code execution and integrates with LLM function calling. \n\nIt also now supports the Model Context Protocol (MCP) for seamless integration with LLM applications.\n\n## Features\n\n- Session-based Python code execution in Docker containers\n- IPython kernel for rich output support\n- Dynamic package installation with security controls\n  - Package allowlist/blocklist system\n  - Version control for security vulnerabilities\n  - Support for pip and conda installations\n- State persistence between executions\n- Support for plotting and visualization\n- Code security validation\n  - AST-based code analysis\n  - Protection against dangerous imports and operations\n  - Support for Jupyter magic commands and shell operations\n- Host directory mounting\n  - Mount local directories into the container\n  - Read-only or read-write access control\n  - Security validations to prevent access to sensitive paths\n\n## MCP Server (Model Context Protocol)\n\nCodeBox-AI now supports the Model Context Protocol (MCP), allowing LLM applications (like Claude Desktop) to interact with your code execution service in a standardized way.\n\n### Running the MCP Server\n\nYou can run the MCP server in several ways:\n\n- **Standalone (for MCP clients or Claude Desktop):**\n  ```bash\n  uv run mcp dev mcp_server.py\n  ```\n  This starts the MCP server in development mode for local testing and debugging.\n\n- **Register with Claude Desktop:**\n  ```bash\n  uv run mcp install mcp_server.py --name \"CodeBox-AI\"\n  ```\n  This will make your server available to Claude Desktop as a custom tool.\n\n- **Combined FastAPI + MCP server:**\n  ```bash\n  uv run run.py\n  ```\n  This starts both the FastAPI API and the MCP server (MCP available at `/mcp`).\n\n- **MCP server only:**\n  ```bash\n  uv run run.py --mode mcp\n  ```\n\n### MCP Fe…","publisher":{"name":"CodeBox AI Team (Thomas Conte and claude.ai)","url":null},"homepage":null,"repository":null,"version":"0.1.0","license":null,"protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:iflow-mcp_tomconte-codebox-ai","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","summary":"pypi","version":"pypi","description":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"code.execution","name":"Code Execution","confidence":1,"provenance":"derived"},{"slug":"dev.terminal","name":"Terminal & Shell","confidence":0.745,"provenance":"derived"}],"categories":["code","dev"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"iflow-mcp_tomconte-codebox-ai","registry":"pypi","observedAt":"2026-09-15T20:21:30.410Z","publishedAt":"2026-02-12T17:59:06.124500Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"iflow-mcp_tomconte-codebox-ai","url":"https://pypi.org/project/iflow-mcp_tomconte-codebox-ai/","firstSeenAt":"2026-09-09T21:24:29.726Z","fetchedAt":"2026-09-15T20:20:13.367Z","normalizedAt":"2026-09-15T20:20:13.367Z"}]},"firstSeenAt":"2026-09-09T21:24:29.726Z","updatedAt":"2026-09-15T20:21:30.410Z"}