oma-scm
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs Git operations (commits, pushes, branch management) using the standard CLI. These operations are governed by safety rules, including the use of --force-with-lease and the requirement for user approval for history rewrites.
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with repository data which could contain malicious instructions.
- Ingestion points: Processes commit messages, diffs, and project history via git status, git diff, and git log.
- Boundary markers: Mandates explicit inspection of staged diffs (git diff --cached) and results before finalizing operations.
- Capability inventory: Has the ability to modify local and remote repository state via git subcommands.
- Sanitization: Instructs the agent to detect sensitive data (secrets/tokens) and stop operations; supports the use of external scanners like gitleaks or trufflehog.
Audit Metadata