skill-eval

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is a developer tool for skill evaluation and demonstrates benign intent.
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to create workspace directories for storing test results.
  • [DATA_EXPOSURE]: The skill reads from and writes to the directory path specified by the user to manage evaluation data and update skill instructions.
  • [PROMPT_INJECTION]: The skill processes content from other skill files and user-defined test prompts. 1. Ingestion points: Reads SKILL.md and evals.json from the user-provided path. 2. Boundary markers: None; content is interpolated into subagent prompts for execution. 3. Capability inventory: Accesses filesystem tools and spawns subagents. 4. Sanitization: Content is used directly as instructions for the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 08:55 PM
Security Audit — agent-trust-hub — skill-eval