pm-backlog-health-auditor
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a structured workflow for auditing backlog data with clear safety mechanisms. It requires the use of a routing guide and 'mutation_gate_v2' to authorize any state changes, effectively preventing unauthorized data modification.
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it analyzes untrusted user content such as backlog story descriptions and acceptance criteria.
- Ingestion points: Reads backlog entities (duplicates, acceptance criteria, hierarchy links) during the audit process.
- Boundary markers: The instructions lack specific delimiters (e.g., XML tags) to isolate analyzed backlog content from the agent's system prompt.
- Capability inventory: The skill can perform 'safe_auto_fixes' on metadata and generate recommendations, though destructive operations are restricted by mandatory confirmation guardrails.
- Sanitization: There are no explicit sanitization routines defined for the data retrieved from external project management tools.
Audit Metadata