commit-push-pr

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform git operations (commit, push, add) and create pull requests using the GitHub CLI (gh). These actions are aligned with the skill's stated purpose of workflow automation.
  • [DYNAMIC_CONTEXT_INJECTION]: Uses shell command substitution (!git branch, !git status, etc.) to prepopulate the agent's context with current branch, status, and diff information when the skill is loaded. These are read-only operations providing necessary context for the task.
  • [DATA_EXPOSURE]: The skill interacts with the local repository's metadata (diffs, logs, status). It does not access sensitive system files or credentials. Network operations (git push, gh pr create) are directed to the user's configured remote repositories.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes git diffs which could potentially contain adversarial instructions embedded in code comments or data. This represents a standard risk surface for tools processing external files, managed here by the LLM's role in synthesizing the information into specific git commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 03:19 PM