glab-mr
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill interacts with the system using
gitandglabcommands. The automated review workflow script (scripts/mr-review-workflow.sh) executes user-supplied test commands but mitigates the risk of arbitrary code execution by validating the input against a predefined allowlist of standard test runners (e.g.,npm test,pytest). - [DATA_EXFILTRATION]: The skill uses GitLab personal access tokens to interact with the GitLab API. These tokens are retrieved securely from environment variables or the
glabconfiguration and are only transmitted to the GitLab host via HTTPS. No patterns of data exfiltration to unauthorized domains were detected. - [PROMPT_INJECTION]: The skill handles untrusted data from GitLab, such as issue titles. The branch creation script (
scripts/create-mr-from-issue.sh) proactively mitigates potential injection risks by sanitizing issue titles—converting them to lowercase and removing non-alphanumeric characters—before using them in shell commands for branch management.
Audit Metadata