triage-issue-local
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains only natural language instructions for repository issue triage. No executable code, remote downloads, or obfuscation techniques were found.
- [DATA_EXFILTRATION]: The instructions include a specific security heuristic to protect user privacy by forbidding the agent from asking for secrets, tokens, or private identifiers in public issue threads.
- [PROMPT_INJECTION]: The skill is designed to process untrusted data from GitHub issues. While this creates a potential surface for indirect prompt injection, the risk is negligible as the skill's capabilities are limited to generating triage questions and applying labels, with no access to sensitive operations.
Audit Metadata