define-domain-language
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: In
SKILL.md, the skill uses agrepcommand to scan the locallib/andapp/directories for Ruby class and module definitions. This is a standard discovery operation used to identify existing terminology in the codebase. - [INDIRECT_PROMPT_INJECTION]: The skill identifies a vulnerability surface for indirect prompt injection because it processes untrusted data from external project documentation.
- Ingestion points: As described in
SKILL.md, the skill collects terms from user-provided requests, PRDs, and tickets. - Boundary markers: There are no specific delimiters or instructions provided to the agent to ignore potentially malicious instructions embedded within the ingested documentation.
- Capability inventory: The skill includes a subprocess call to
grep(documented inSKILL.md) but does not possess high-risk capabilities such as network access, arbitrary file writing, or administrative privilege requirements. - Sanitization: No sanitization or validation of the external documentation content is specified before it is processed by the agent.
Audit Metadata