flexport-deploy-integration

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides templates for processing untrusted external data via webhook receivers.
  • Ingestion points: The skill defines webhook payload processing routes in app/api/webhooks/flexport/route.ts (SKILL.md).
  • Boundary markers: The provided code implementation includes mandatory HMAC signature verification using crypto.createHmac and crypto.timingSafeEqual to validate request authenticity before processing.
  • Capability inventory: The skill utilizes Bash for cloud deployments (via Fly.io and Google Cloud Run) and Write/Edit for configuration management.
  • Sanitization: Instructions explicitly mandate the exclusion of credentials, shipment payloads, addresses, and invoices from logs, reducing the risk of data leakage.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 08:37 AM
Security Audit — agent-trust-hub — flexport-deploy-integration