Skills
skills/0xbigboss/claude-code/git-best-practices/Gen Agent Trust Hub

git-best-practices

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill ingests untrusted data from repository states via git status and git diff. Evidence: 1. Ingestion points: git status and git diff HEAD (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: git add, git commit, git push (SKILL.md). 4. Sanitization: Absent. The risk is low as the data influence is limited to structural git operations.
  • [Data Exposure] (SAFE): Explicitly instructs the agent to avoid committing secrets, credentials, and .env files.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 07:28 AM