skills/tuesd4y/agent-skills/commit/Gen Agent Trust Hub

commit

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes Git commands and a bundled context script. * Script: scripts/gather-context.sh collects branch names and commit history. * Scope: Tool usage is restricted to standard Git operations and specific bundled scripts. * Control: A mandatory approval step using AskUserQuestion is enforced before any permanent changes or commits are made.
  • [PROMPT_INJECTION]: The skill processes file diffs and hook outputs which are externally controlled (Indirect Prompt Injection). * Ingestion points: The agent reads git diff output and pre-commit hook failure messages. * Capability inventory: Includes the ability to stage files and edit code to fix formatting issues. * Sanitization: The git commit command uses a heredoc pattern to ensure commit messages are treated as literal data rather than executable shell code. * Boundary markers: Uses a human-in-the-loop confirmation process for all commit actions and substantive code fixes.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 03:17 PM
Security Audit — agent-trust-hub — commit