commit-changes

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several Git commands (git branch, git status, git diff, git add, git commit, git log) to manage source control. These operations are restricted to the local repository and are consistent with the skill's stated purpose of automating commits.
  • [DYNAMIC_CONTEXT_INJECTION]: The skill uses the !command`` syntax in SKILL.md to dynamically inject the current branch name, status summary, and diff statistics into the agent's context. The commands used (git branch --show-current, git status --short, and git diff --stat HEAD) are benign and do not incorporate unvalidated user input, making them safe for pre-execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local environment by reading git diff HEAD. While this represents a potential injection surface where malicious instructions could be embedded in code comments or diff content, the skill's capabilities are limited to Git operations and do not include high-risk actions like network exfiltration or arbitrary code execution, resulting in a safe risk profile for its intended use case.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 03:18 PM
Security Audit — agent-trust-hub — commit-changes