atomic-commits-philosophy
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- Indirect Prompt Injection (LOW): The skill instructs the agent to read
git log --oneline -10to determine commit style, which is a tool output poisoning surface (Category 8c). 1. Ingestion points: SKILL.md (via git log command). 2. Boundary markers: Absent; no instructions are provided to ignore embedded commands within the log history. 3. Capability inventory: The skill is intended for agents with git and filesystem access. 4. Sanitization: Absent. - NO_CODE (SAFE): The skill contains no executable scripts, binaries, or active code components.
Audit Metadata