khe-brain-yc
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates on a provided set of Markdown files containing paraphrased advice. It does not request excessive permissions or perform operations outside of its stated educational scope.
- [EXTERNAL_DOWNLOADS]: All external URL references are to well-known and authoritative startup resources (ycombinator.com and paulgraham.com) directly related to the skill's primary purpose.
- [COMMAND_EXECUTION]: The skill uses a local script (
scripts/brain-check.mjs) to verify the AI's output. The script is implemented using standard Node.js modules, contains no external dependencies, and is limited to regex-based text analysis of local files. It does not perform network calls or execute dynamic content. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied startup plans but manages the risk of instruction override through a rigorous workflow, strict formatting rules, and an automated verification gate that checks every citation against an authoritative source list.
Audit Metadata