work-items-breakdown
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes official command-line tools including
gh,glab, andazto interact with GitHub, GitLab, and Azure DevOps for task management. These operations are limited to checking authentication status and creating work items. - [INDIRECT_PROMPT_INJECTION]: The skill processes content from local project files like
docs/<slug>-spec.mdanddocs/<slug>-plan.md. While these files are external inputs, the skill implements a mandatory human-in-the-loop process requiring a full preview and explicit user confirmation before any items are created. 1. Ingestion points: Reads local markdown specification and planning documents. 2. Boundary markers: Instructions explicitly mandate showing a preview and obtaining a clear 'yes' from the user. 3. Capability inventory: File writing for local task lists and execution of tracker-specific CLI commands. 4. Sanitization: The skill relies on user review rather than automated content sanitization.
Audit Metadata