push
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill exclusively provides instructional content for Git operations. All provided commands are standard Git CLI operations intended for local and remote repository synchronization.
- [CREDENTIALS_UNSAFE]: The skill correctly identifies the risks associated with hardcoded credentials and advocates for the use of secure alternatives like SSH keys, personal access tokens, and credential helpers. It provides safe troubleshooting steps for SSH authentication that align with industry best practices.
- [COMMAND_EXECUTION]: While the skill heavily uses shell commands (
git,ssh,gh), these are within the expected scope of a Git management skill. No arbitrary or suspicious command execution patterns were detected. - [DATA_EXFILTRATION]: There are no indicators of data exfiltration. All network operations (e.g.,
git push,ls-remote) are directed at user-configured remote repositories or well-known services like GitHub. - [PROMPT_INJECTION]: The instructions do not contain patterns attempting to override agent behavior or bypass safety filters. The language is instructional and task-oriented.
Audit Metadata