commit
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points: The skill instructs the agent to read and follow content from the local file
.ai-engineering/skills/commit/SKILL.md(SKILL.md). - Boundary markers: No delimiters or instructions are provided to distinguish between the file's content and the agent's internal instructions.
- Capability inventory: The description suggests the agent will perform tasks like staging, linting, and pushing commits, which typically involve command line execution and file system access (SKILL.md).
- Sanitization: No validation or sanitization is performed on the content of the referenced file before it is executed or processed.
Audit Metadata