fix-ci

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use standard GitHub CLI (gh) commands to view pull request checks and run logs. This is essential for its functioning and poses no inherent security risk.- [PROMPT_INJECTION]: The skill processes job output and logs which could contain content from external sources (e.g., code in pull requests). This represents a surface for indirect prompt injection, though no malicious behavior is directed by the skill itself.\n
  • Ingestion points: CI logs via gh run view --log-failed.\n
  • Boundary markers: None.\n
  • Capability inventory: Local command execution and file system access.\n
  • Sanitization: Not specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 11:09 AM
Security Audit — agent-trust-hub — fix-ci