flexport-cost-tuning

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides legitimate architectural guidance and TypeScript code snippets to reduce API call volume for Flexport integrations. These examples follow standard industry practices for cost optimization and API monitoring.\n- [INDIRECT_PROMPT_INJECTION]: The skill documents a webhook-driven architecture that ingests data from external API endpoints, creating a potential surface for untrusted data processing.\n
  • Ingestion points: The TypeScript example in SKILL.md defines a webhook listener at /webhooks/flexport to receive shipment milestones.\n
  • Boundary markers: No specific delimiters or validation logic are included in the simplified architectural code examples.\n
  • Capability inventory: The skill is configured with Read, Write, and Edit tool permissions, which allow for code modifications related to these optimizations.\n
  • Sanitization: The example logic updateDB(event.data) processes the raw request body without demonstrating explicit sanitization or validation steps within the documentation snippet.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 08:37 AM
Security Audit — agent-trust-hub — flexport-cost-tuning