idea-finder
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is primarily designed to orchestrate the
idea-finderCLI tool to perform demand discovery, workspace diagnostics, and validation tracking. It uses specific command-line arguments and requires the use of absolute workspace paths to ensure predictable execution. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests and summarizes qualitative demand data from untrusted external sources (e.g., HN, V2EX, App Store reviews, Stack Exchange).
- Ingestion points: Research results are pulled from public web sources and manual imports from local files like
fetched.jsoninto the agent's context. - Boundary markers: The instructions mandate citing specific quotes and provenance, which assists in separating source material from agent reasoning, though technical delimiters are not explicitly defined.
- Capability inventory: The agent can execute CLI commands that perform network collection and local file system operations within the research workspace.
- Sanitization: No specific sanitization or filtering logic is provided for the text retrieved from external qualitative sources before it is processed by the agent.
Audit Metadata