cap-commit-push
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automates Git operations by executing shell commands such as
git status,git add -A,git commit, andgit pushas defined inSKILL.md. - [PROMPT_INJECTION]: The skill generates commit messages based on the current task context, creating a surface for indirect prompt injection if the project content is untrusted. 1. Ingestion points: Task context (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Shell command execution (SKILL.md). 4. Sanitization: Absent.
- [SAFE]: The skill performs expected developer actions within the local repository and does not exhibit patterns of credential exfiltration, obfuscation, or unauthorized remote code execution.
Audit Metadata