khe-brain-gary
Pass
Audited by Gen Agent Trust Hub on Nov 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a structured knowledge base with high-precision citation requirements and no detected malicious intent.
- [COMMAND_EXECUTION]: Includes a Node.js utility script (
scripts/brain-check.mjs) used by the agent to perform static analysis on its own generated responses. This script is restricted to reading local files within the skill directory and does not contain network access or dynamic evaluation logic. - [INDIRECT_PROMPT_INJECTION]: The skill defines a strict workflow for processing user queries against its internal corpus. It uses backticks as delimiters for user input and employs the aforementioned validation script to ensure the agent's response adheres to citation rules and does not exhibit unauthorized behavior resulting from untrusted data ingestion.
Audit Metadata