ln-130-tasks-docs-creator

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill handles documentation files which represent an indirect prompt injection surface.
  • Ingestion points: docs/tasks/README.md and docs/tasks/kanban_board.md which are read and validated for content.
  • Boundary markers: Utilizes SCOPE tags (e.g., <!-- SCOPE: ... -->) and specific markdown headers to identify and constrain processed content.
  • Capability inventory: Operates with file system capabilities including Read, Write, Edit, Glob, and mkdir for document management.
  • Sanitization: Employs strict regex validation for user-provided data like Linear Team UUIDs and Team Keys.
  • [EXTERNAL_DOWNLOADS]: Fetches the Mermaid.js library from a well-known CDN (cdn.jsdelivr.net) within the diagram.html file for state diagram rendering.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 11:56 PM