functional-planner
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill is designed to ingest and process external data from Product Requirements Documents (PRDs) and project management platforms, which serves as a potential vector for indirect prompt injection.
- Ingestion points: PRD source artifacts (files, folders, or explicit content), codebase context, and read-only metadata from Azure DevOps, GitHub, and Jira trackers.
- Boundary markers: The skill documentation includes an explicit "Untrusted Content Boundary" rule intended to prevent requirement text from overriding system behavior or triggering unauthorized mutations.
- Capability inventory: The skill performs file system writes to maintain planning logs and generate handoff artifacts, and executes read-only tool calls to external tracker APIs for validation.
- Sanitization: The skill references a requirement for content sanitization (delegated to the backlog-management dependency) before data is presented for user review or further processing.
Audit Metadata