distilling-domain-language
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied terminology and scenarios during the 'Candidate dump' and 'Scenario ambush' phases. It also reads content from the repository and
CONTEXT.mdviagrep. While this provides an attack surface for indirect prompt injection if the source files or user inputs contain malicious instructions, the skill uses this data for a specific modeling purpose without executing it as code. - [COMMAND_EXECUTION]: The skill instructs the agent to 'Grep the repo' to find synonym clusters. This is a standard utility command for exploring codebases and does not involve high-risk operations or privilege escalation.
Audit Metadata