Skills
skills/malikasadjaved/skills/fastapi-best-practices/Gen Agent Trust Hub

fastapi-best-practices

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill defines several API endpoints that ingest untrusted user data via Pydantic models, creating a surface for indirect prompt injection. Ingestion points: SKILL.md (DocumentCreate model) and scripts/app_template.py (ItemCreate model). Boundary markers: The skill uses Pydantic validation (e.g., min_length, max_length, and pattern) to define input constraints. Capability inventory: Associated capabilities across the provided scripts include database interactions (db.execute in SKILL.md), in-memory storage (_items in scripts/app_template.py), and outbound network requests (httpx in SKILL.md and scripts/dependencies.py). Sanitization: Sanitization is demonstrated through Pydantic type checking and field-level validation logic.
  • [EXTERNAL_DOWNLOADS]: The skill provides installation commands for well-known and standard Python libraries such as fastapi, pydantic, httpx, and uvicorn from official registries.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 03:40 PM
Security Audit — agent-trust-hub — fastapi-best-practices