idea-intake
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill manages files within the user's workspace, specifically creating stage documentation (e.g., 00-idea-intake.md) and updating project metadata (status.json). These operations are restricted to the local workspace and follow a predefined workflow.
- [EXTERNAL_DOWNLOADS]: The skill utilizes web search to research unknown concepts provided in user ideas. This functionality is governed by a clear research policy, requires the creation of a research plan, and stores results in structured JSON and markdown files.
- [PROMPT_INJECTION]: While the skill processes untrusted user input (raw ideas) and external web data, it effectively mitigates indirect prompt injection risks by employing strict markdown templates for output and JSON schema validation for source and status tracking.
Audit Metadata