fastapi-async-patterns

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The provided code snippets contain placeholder credentials in connection strings, such as "user:pass" in "postgresql+asyncpg://user:pass@localhost/db" and "mongodb://localhost:27017".
  • [EXTERNAL_DOWNLOADS]: Code examples include HTTP requests to placeholder domains (e.g., "api.example.com", "store1.com", "store2.com") for instructional purposes.
  • [INDIRECT_PROMPT_INJECTION]: The skill demonstrates handling untrusted input (e.g., filenames from uploads, WebSocket data) that is used in capabilities like file system access and database queries without demonstrating sanitization.
  • Ingestion points: Data enters via the "filename" parameter in the "/download/{filename}" route and the "file.filename" in the "/upload" route within SKILL.md.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the examples.
  • Capability inventory: Scripts demonstrate file reading/writing via "open" and file deletion via "os.remove" in the "cleanup_temp_files" task.
  • Sanitization: No path validation or input sanitization logic is included in the tutorial snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:09 PM
Security Audit — agent-trust-hub — fastapi-async-patterns