skills/mryll/skills/pr-quiz/Gen Agent Trust Hub

pr-quiz

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration techniques were detected. The skill operates within the expected scope of a developer assistant tool.
  • [DATA_EXPOSURE]: The skill accesses code diffs using git diff to generate relevant questions. This is necessary for its primary purpose and occurs within the local repository environment to analyze changes.
  • [PERSISTENCE]: The skill updates a local markdown file (comprehension map) to record user progress. This persistence is a documented feature and does not involve system-level modifications or hidden startup mechanisms.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from code diffs and user responses. While this represents a data ingestion surface, the instructions specifically mandate grounding questions in the actual code diff and strictly separate the agent's grading role from the user's role, reducing the likelihood of successful manipulation via external content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 08:00 PM
Security Audit — agent-trust-hub — pr-quiz