make-github-issues
Pass
Audited by Gen Agent Trust Hub on May 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted content from the codebase to generate issues.
- Ingestion points: codebase files and directories processed by the agent (SKILL.md).
- Boundary markers: absent; no delimiters or instructions are provided to distinguish codebase content from agent instructions.
- Capability inventory: uses 'gh issue create' for network-bound issue reporting and writes a findings report to the local file system.
- Sanitization: absent; the skill does not validate or sanitize code findings before including them in external issue descriptions.
Audit Metadata