ios-commit

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes standard git commands such as status, diff, and commit to facilitate version control tasks within the local environment.
  • [SAFE]: Explicit pre-commit checks are included to warn users against committing sensitive information, such as credentials or environment files.
  • [SAFE]: The workflow incorporates human-in-the-loop verification by requiring user confirmation for generated messages and execution of the commit command.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it analyzes external data from git diffs and project files. 1. Ingestion points: staged file changes and the project's CLAUDE.md file. 2. Boundary markers: No explicit delimiters are used to separate untrusted code diffs from instructions. 3. Capability inventory: Command execution via the git CLI. 4. Sanitization: No sanitization of analyzed content prior to commit message generation is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 01:51 AM
Security Audit — agent-trust-hub — ios-commit