to-issues

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses the official GitHub CLI tool (gh) for viewing and creating issues, which is standard for developer-oriented agents.
  • [SAFE]: A mandatory human-in-the-loop review process (Step 4) is implemented, requiring user approval for all proposed issues before they are published. This is an effective security practice to prevent unauthorized or unintended modifications to the issue tracker.
  • [SAFE]: The skill processes data from external sources (existing GitHub issues and comments) but mitigates potential indirect prompt injection by ensuring the agent drafts content for user verification rather than executing commands automatically based on that data.
  • [SAFE]: The frontmatter includes disable-model-invocation: true, which restricts the agent's ability to trigger further model calls, narrowing the attack surface and preventing complex multi-step automated behaviors.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 06:58 AM
Security Audit — agent-trust-hub — to-issues