domain-modeling
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill contains a standard attack surface for indirect prompt injection as it is designed to process untrusted project data.
- Ingestion points: The skill is instructed to cross-reference user statements with codebase content in
SKILL.md. - Boundary markers: No explicit boundaries or delimiters are defined to separate untrusted code content from the agent's instructions.
- Capability inventory: The skill has the capability to write and update Markdown files (CONTEXT.md, CONTEXT-MAP.md, and ADRs) within the repository.
- Sanitization: There is no mention of sanitizing or escaping content read from the codebase before it is used to generate or update documentation.
- [NO_CODE]: The skill consists solely of Markdown documentation and YAML metadata. It does not ship with scripts, binary files, or instructions to execute arbitrary shell commands.
Audit Metadata