MCP Tool Contract Builder

WHY THIS EXISTS

Built for AI-agent work that needs evidence, not vibes.

Draft an MCP-style tool contract with input schemas, permission boundaries, eval cases, abuse tests, rate limits, and agent handoff documentation. The useful output is a visible table plus a receipt that names input, checks, limits, and next action.

• Forces every tool to declare scope, inputs, outputs, errors, side effects, and approval gates.
• Creates positive evals plus abuse cases instead of only a happy-path manifest.
• Gives AI agents a safer handoff before connecting real tools.

Boundary: Not a guarantee of MCP compliance, production security review, legal terms, or a substitute for testing the real server implementation.