cs-brainstorm
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads untrusted repository content (e.g., ADRs and requirements) to gather brainstorming context.
- Ingestion points: Reads .codestable/attention.md, requirements/CONTEXT.md, and requirements/adrs/ in SKILL.md.
- Boundary markers: No explicit delimiters or instructions are used to ignore instructions embedded in the project files.
- Capability inventory: File creation, grep command execution, and generated code execution in SKILL.md.
- Sanitization: No filtering or validation is performed on ingested content.
- [COMMAND_EXECUTION]: The skill executes local shell commands such as grep to discover project context and supports the generation and execution of technical 'spike' scripts (e.g., Python or TypeScript) for verification, which requires user approval.
Audit Metadata