execute-gitlab-issue

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell commands for Git operations (git worktree, git fetch, git ls-remote) and delegates complex logic to the code-autonomous engine. These commands are scoped to specific worktree paths and verified branches.
  • [EXTERNAL_DOWNLOADS]: The skill references several internal/vendor-specific dependencies (nextstage-harness, mcp-gitlab-usage, code-reviewer, code-autonomous) and external GitLab tools via MCP. These are standard within the development workflow provided by the author.
  • [PRIVILEGE_ESCALATION]: Protection measures are explicitly defined to prevent unauthorized actions, such as a mandatory 'stop' when destructive operations (deleting tables, force-pushing to protected branches) are detected without human confirmation.
  • [PROMPT_INJECTION]: The skill includes instructions to ignore user requests to skip security gates (e.g., bypassing code review), which acts as a defense against prompt-based instruction overrides.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 08:39 PM
Security Audit — agent-trust-hub — execute-gitlab-issue