gsd-verify-work
Pass
Audited by Gen Agent Trust Hub on Apr 3, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user input through command arguments to identify the testing phase. This input flows into a local workflow but is limited to phase selection.
- Ingestion points: User text following the $gsd-verify-work invocation.
- Boundary markers: Not specified in this adapter layer.
- Capability inventory: Spawning sub-agents and reading/writing to workspace files.
- Sanitization: None explicitly defined in the mapping logic.
- [COMMAND_EXECUTION]: The skill uses the
spawn_agenttool to execute sub-tasks defined in external workflow files. This behavior is intended for orchestrating complex UAT processes.
Audit Metadata