gsd-next
Pass
Audited by Gen Agent Trust Hub on Apr 3, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external project files to determine its next actions.
- Ingestion points: Reads project status information from
STATE.mdandROADMAP.md. - Boundary markers: Absent; the skill does not use delimiters to distinguish project data from its own instructions.
- Capability inventory: Can trigger
spawn_agentandrequest_user_inputbased on the parsed data. - Sanitization: No content filtering or validation is performed on the ingested data.
- [COMMAND_EXECUTION]: The skill is configured to automatically execute workflow instructions from the local path
@/mnt/local-analysis/workspace-hub/.codex/get-shit-done/workflows/next.md. While this is the intended functionality, it involves the dynamic execution of logic stored in a file that could potentially be modified.
Audit Metadata