api-connector

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides instructional content and code templates for integrating with reputable and well-known service providers such as Stripe, OpenAI, Anthropic, and Supabase. All network targets in the examples are official API endpoints. \n- [PROMPT_INJECTION]: The Next.js API route templates in SKILL.md demonstrate a surface for indirect prompt injection where user-supplied data is interpolated directly into model messages without explicit guards. \n
  • Ingestion points: Data is ingested from request.json() and req.body in the provided API route examples within SKILL.md. \n
  • Boundary markers: No delimiters or instructions to ignore embedded commands are present in the examples. \n
  • Capability inventory: The examples utilize outbound network requests to communicate with external AI providers. \n
  • Sanitization: No input validation, filtering, or escaping logic is demonstrated in the provided code snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 05:04 AM
Security Audit — agent-trust-hub — api-connector