commit-staged
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is entirely instructional and does not execute scripts, make network requests, or access sensitive files. It defines a workflow for committing changes using existing project styles and a secondary rule-loading skill.
- [INDIRECT_PROMPT_INJECTION]: The skill relies on analyzing existing commit history to determine message style, which is an untrusted ingestion point. This represents a minor surface for indirect prompt injection if repository history contains malicious content designed to influence the agent. 1. Ingestion points: Project commit history (referenced in SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Git commit generation and execution. 4. Sanitization: None specified in the instructions.
Audit Metadata