tdd
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes shell commands such as
git diffandgit logto provide context on code changes and test failures. - [COMMAND_EXECUTION]: Recommends the use of the
helpmetestcommand-line tool for execution when platform-integrated tools are unavailable. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. Ingestion points: Reads agent personality and instructions from
.helpmetest/SOUL.mdand reads task state from artifacts. Boundary markers: Absent; there are no delimiters or instructions to ignore commands within the ingested content. Capability inventory: Accesses shell commands (git) and testing utilities with system interaction. Sanitization: Absent; the skill does not validate or filter content from the repository files before integrating it into the agent's context.
Audit Metadata