gh-issue-report
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: Uses the
ghcommand-line tool to interact with GitHub repositories, which is appropriate for its described functionality. - [DATA_EXFILTRATION]: All data handled by the skill is used to generate bug reports on GitHub. There are no patterns suggesting data is being sent to external or unauthorized servers.
- [PROMPT_INJECTION]: While the skill ingests external data from GitHub (issue templates and code), the potential for indirect prompt injection is mitigated by the design. * Ingestion points: Repository content and issue templates fetched via the GitHub API in SKILL.md. * Boundary markers: Absent. * Capability inventory: Ability to create issues on GitHub using
gh issue createas defined in SKILL.md. * Sanitization: Not explicitly performed on external data; however, the skill mandates a user approval step in Step 13 before any content is submitted, effectively preventing automated malicious actions.
Audit Metadata