skills/utarn/engineer-skills/qa/Gen Agent Trust Hub

qa

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI tool (gh issue create) to create new issues. This is the core functionality of the skill and is used as intended for issue tracking.
  • [DATA_EXFILTRATION]: The skill transmits project-related information and user-reported issues to GitHub. The instructions include privacy-focused constraints, such as forbidding the inclusion of file paths, line numbers, or internal implementation details, to ensure that only behavioral descriptions are shared externally.
  • [PROMPT_INJECTION]: The skill processes untrusted data from user conversations and potentially malicious content within the codebase (e.g., UBIQUITOUS_LANGUAGE.md) which represents an indirect prompt injection surface.
  • Ingestion points: User chat inputs and project documentation files read during the exploration phase.
  • Boundary markers: The skill employs specific Markdown templates to structure the output sent to the GitHub CLI.
  • Capability inventory: Execution of the GitHub CLI (gh) for remote resource creation.
  • Sanitization: The instructions mandate focusing on user-facing behaviors and project-specific domain language rather than raw code or technical internals.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 05:07 AM
Security Audit — agent-trust-hub — qa