triage
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from external sources (issue bodies, comments, and pull request diffs).
- Ingestion points: As defined in
SKILL.md, the agent gathers data from issue bodies, comments, and PR diffs during the triage process. - Boundary markers: The instructions do not specify the use of delimiters or explicit warnings to the agent to ignore embedded instructions within the processed data.
- Capability inventory: The skill allows the agent to modify labels, post comments to the tracker, search the codebase, and create files in the
.out-of-scope/directory. - Sanitization: No explicit validation or sanitization of the gathered external data is defined in the instructions.
Audit Metadata