Agent♥︎Age
Catalog

IntentFence Agent Action Firewall

Official10 toolsLive

by razel369 · TypeScript

Fail-closed action authorization, MCP risk scanning, x402 checks, and signed receipts.

io.github.razel369/intentfence — Model Context Protocol (MCP) Security & Authorization Server

IntentFence is a fail-closed authorization firewall for autonomous AI agents. It performs MCP risk scanning and x402 checks, and validates action execution. Immediately before a consequential MCP, HTTP, A2A, or payment action, it binds the exact action and payload hash to explicit action/resource allowlists, spend and retention ceilings, and optional action-specific approval.

🛠️ Key Features

  • Fail-closed action authorization
  • MCP risk scanning
  • x402 checks
  • Action/resource allowlists with spend and retention ceilings
  • Five-minute ES256 receipts; TypeScript and Python SDK verification

🚀 Use Cases

  • Gating consequential MCP actions
  • Controlling HTTP, A2A, and payment actions via allowlists and ceilings
  • Verifying caller-owned action digests before executing callbacks

⚡ Developer Benefits

  • Receipt-based verification using ES256
  • SDKs verify the receipt and local action digest before execution
  • Provides MCP metadata

⚠️ Limitations

  • Never executes downstream tools
  • Should not receive credentials or wallet keys

Topics

a2aagentic-aiai-agentsmcppolicy-as-codesecurityx402agent-paymentsgemini-cli-extensionpayment-firewall

Related servers

More in Security
IntentFence Agent Action Firewall - agentage MCP Catalog