tool-discovery

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides a set of static references for tool selection. The catalogs map common development tasks to specific libraries and agents.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing missing dependencies via pip or uv. The suggested packages (e.g., pandas, numpy, requests, tensorflow) are industry-standard, well-known libraries from the official Python Package Index.
  • [PROMPT_INJECTION]: The skill identifies tools based on user-provided task descriptions, creating a surface for indirect prompt injection.
  • Ingestion points: User-provided task descriptions used for discovery in SKILL.md.
  • Boundary markers: None present to distinguish user input from system instructions.
  • Capability inventory: The recommended tools include those for network operations, file system access, and package installation.
  • Sanitization: No input validation or sanitization is performed before matching against triggers.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 05:05 PM