triage-issue
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists of natural language instructions and metadata with no executable code or scripts.
- [PROMPT_INJECTION]: The skill's workflow presents a surface for indirect prompt injection as it processes untrusted data from GitHub issues. Ingestion points: External GitHub issue content. Boundary markers: Absent. Capability inventory: Read-only GitHub and local repository access. Sanitization: None.
- [SAFE]: All described operations are consistent with the skill's purpose of issue triaging and do not involve data exfiltration or unauthorized system access.
Audit Metadata