review-domain-boundaries
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill recommends using
rg(ripgrep) to search for project-specific business terms and namespace usage across source directories. These operations are read-only and performed locally. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes existing project source code, which represents an ingestion point for external data that could contain adversarial prompts.
- Ingestion points: Application source files within the
lib/,app/, andservices/directories, as specified inSKILL.md. - Boundary markers: No specific delimiters or instructions to ignore embedded prompts are defined for the code analysis process.
- Capability inventory: The skill is restricted to searching and reading local project files; it lacks instructions or capabilities for network access, file writing, or administrative command execution.
- Sanitization: No sanitization or content validation of the project source code is described.
Audit Metadata