sdd-audit
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill includes a shell script
scripts/validate.shthat executes local commands includinggrep,find, andnpx. These are used for static analysis and running standard development tools (TSC, ESLint). - [INDIRECT_PROMPT_INJECTION] (LOW): The workflow involves reading external documentation files (
spec.md,plan.md, etc.). While this is an ingestion surface for untrusted data, the impact is limited to the generated audit report. - Ingestion points:
spec.md,plan.md,tasks.md,todo-list.md(read via file system) - Boundary markers: None explicitly defined in the workflow for separating data from instructions.
- Capability inventory: Local file system read access and shell command execution via
scripts/validate.sh. - Sanitization: None; the agent is expected to interpret the content of these files directly.
Audit Metadata