skills/bastani-inc/atomic/gh-commit/Gen Agent Trust Hub

gh-commit

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes dynamic context injection (the !command syntax) to execute local Git utilities such as git status, git branch, git diff, and git log. These commands are used solely to provide the AI agent with context about the repository's state (staged files, current branch, and recent history). No user-supplied input is passed to these commands, and they are restricted to read-only operations on repository metadata.
  • [PROMPT_INJECTION]: The skill contains instructional directives such as 'IMPORTANT: DO NOT SKIP pre-commit checks' and 'ALWAYS attribute AI-Assisted Code Authorship'. These are categorized as standard procedural instructions for maintaining project standards and do not attempt to override the AI agent's core safety guardrails or system instructions.
  • [DATA_EXFILTRATION]: While the skill accesses source code diffs to help generate commit messages, it does not include any network-capable commands (e.g., curl, wget) or instructions to transmit the gathered data to external servers. All operations are confined to the local environment and are consistent with the skill's stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 03:26 PM
Security Audit — agent-trust-hub — gh-commit