ln-302-task-replanner
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious behavior detected. The skill uses authorized internal reference files and follows intended synchronization logic.
- [PROMPT_INJECTION]: The skill includes an attack surface for indirect prompt injection through its core task-processing features.
- Ingestion points: Story data (id/title/description) and existing task descriptions from Linear (SKILL.md).
- Boundary markers: No explicit instruction-separation delimiters are used.
- Capability inventory: Modifies Linear issue states and descriptions, and updates local kanban_board.md files.
- Sanitization: Performs mandatory scanning for destructive operation keywords using shared safety templates.
Audit Metadata