cs-feat-design
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute local scripts and CLI tools as part of the design and validation workflow. Evidence includes instructions to run 'python .codestable/tools/validate-yaml.py' and 'python .codestable/tools/search-yaml.py', as well as references to tools like 'cs-onboard' and 'cs-roadmap'.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection (Category 8). Ingestion points: User-provided feature slugs, summary text, and the content of intent ({slug}-intent.md) or brainstorm documents. Boundary markers: No explicit delimiters or warnings are used when interpolating intent or brainstorm content. Capability inventory: File system operations (directory creation, document writing) and execution of local Python scripts. Sanitization: Instructions include a basic constraint for the 'slug' variable (lowercase alphanumeric and hyphens), but intent content is not sanitized before processing.
Audit Metadata