triage-frontend-issues
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from Sentry issues (titles, stack traces) which constitutes an ingestion surface for untrusted content. However, this risk is mitigated by a mandatory human-in-the-loop approval process for the triage plan and restricted tool capabilities. 1. Ingestion points: search_issues and get_sentry_resource results (SKILL.md). 2. Boundary markers: Explicit plan presentation and user 'apply' confirmation (SKILL.md Step 3 and 4). 3. Capability inventory: update_issue limited to status='ignored'. 4. Sanitization: Human verification of categories and reasons before tool execution.
- [SAFE]: The skill follows least-privilege principles by restricting operations to a specific project and organization, and only permitting the archive (ignored) status mutation. All remote resources (mcp.sentry.dev) are official and trusted services.
Audit Metadata