workflow-feedback-to-closure
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external data (user reports, reviews, Sentry issues) which creates a surface for indirect prompt injection. This risk is inherent to the feedback triage function. \n
- Ingestion points: Phase 0 (SKILL.md) reads user reports, support messages, app-store reviews, Sentry issues, and QA findings. \n
- Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are mentioned in the normalization step. \n
- Capability inventory: The skill utilizes
gh issue create(Phase 2), triggers other workflow skills (Phase 4), and executestest-playwright(Phase 5). \n - Sanitization: No specific sanitization or validation of the raw incoming feedback signals is described. \n- [NO_CODE]: The skill consists exclusively of instructional Markdown text and YAML configuration; it does not ship with or download executable code or scripts.
Audit Metadata