ln-513-auto-test-planner
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill ingests untrusted data from Linear comments to generate automated test plans. (1) Ingestion points: External Linear comments (Manual Testing Results and Test Research) fetched in Phase 2. (2) Boundary markers: Relies on specific markdown headers and format version strings (e.g., '## Manual Testing Results'). (3) Capability inventory: Triggers task creation or modification via delegation to ln-301-task-creator or ln-302-task-replanner. (4) Sanitization: Employs regex-based parsing and structural validation of sections, though these do not fully prevent instruction injection within values.
- Data Exposure (SAFE): The skill performs local discovery by reading project configuration files (e.g., jest.config.js) and task documentation (kanban_board.md) to align planning with project practices.
Audit Metadata