Mathematical Code Execution

Mathematical Code Execution

A Python-only sandbox for agent hosts. The model writes Python that prints a result (or a one-page PDF via preinstalled fpdf); this server runs it in an isolated process with no network, no user filesystem and no package installs. Other languages are refused. numpy is available; scipy and pandas are not.

Registry name	ai.saifs/code-execution
Endpoint	https://code-execution-mcp.saifs.ai/mcp
Transport	Streamable HTTP (remote, nothing to install)
Auth	OAuth 2.1 — PKCE (S256) and dynamic client registration; your client opens a sign-in page on first use
Publisher	SAIFS AI

Install

VS Code / GitHub Copilot

Add to .vscode/mcp.json (workspace) or your user mcp.json:

{
  "servers": {
    "saifs-code-execution": {
      "type": "http",
      "url": "https://code-execution-mcp.saifs.ai/mcp"
    }
  }
}

Then start the server from the MCP view and sign in when prompted.

Claude (claude.ai, Desktop)

Settings → Connectors → Add custom connector, and paste:

https://code-execution-mcp.saifs.ai/mcp

Claude Code

claude mcp add --transport http saifs-code-execution https://code-execution-mcp.saifs.ai/mcp

Any other MCP client

Point it at https://code-execution-mcp.saifs.ai/mcp using the Streamable HTTP transport. The server returns 401 with a WWW-Authenticate: Bearer … resource_metadata=… challenge, so OAuth-aware clients sign in automatically.

Tools

Tool	What it does
execute_code	Runs allow-listed Python and returns stdout, stderr, exit code, and any PDF written with fpdf (output("name.pdf"), download link expires in 1 hour). Non-Python is rejected with a one-line warning.

Example prompts

• “Run Python to compute the mean and standard deviation of this list and print the result.”
• “Write a short numpy snippet that multiplies these two matrices and prints the product.”
• “Generate a one-page PDF invoice summary with fpdf and give me the download link.”

Support

• Website: https://saifs.ai
• Issues: https://github.com/Saify-Technologies/saifs-ai-mcps-docs/issues
• Privacy: https://saifs.ai/privacy_policy