NYC

FastAPI Customer Support Tech Enablement

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION] (HIGH): The skill is designed for Indirect Prompt Injection (Category 8) vulnerability.
  • Ingestion points: Processes untrusted external content via Support Ticket CRUD, WebSocket real-time chat, and file uploads (EXAMPLES.md).
  • Boundary markers: No explicit evidence of prompt delimiters or instructions to ignore embedded commands in the summary.
  • Capability inventory: Includes database modification (SQLAlchemy), file system storage (File Uploads), and external communications (Background Tasks/Email).
  • Sanitization: Summary does not mention specific sanitization or filtering logic for incoming customer messages.
  • [EXTERNAL_DOWNLOADS] (LOW): Requires installation of several Python packages (fastapi, uvicorn, sqlalchemy, asyncpg, pydantic, python-jose, passlib). These are standard, widely-trusted ecosystem packages from PyPI.
  • [COMMAND_EXECUTION] (LOW): Instructions include standard development commands like 'pip install' and 'uvicorn filename:app --reload'.
  • [INFO] (LOW): Automated scan flagged 'logger.info' as a malicious URL. This is identified as a false positive where the scanner misinterprets Python logging syntax as a malicious domain string.
Recommendations
  • AI detected serious security threats
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 12:01 AM