finishing-a-development-branch

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION] (SAFE): The skill executes local shell commands including git, npm test, pytest, cargo test, and go test. These operations are essential to its primary purpose of verifying code and managing development workflows.
  • [DATA_EXPOSURE & EXFILTRATION] (SAFE): No unauthorized file access or data exfiltration attempts were detected. Network activity is limited to standard Git operations (git push) and GitHub Pull Request creation via the gh CLI.
  • [INDIRECT PROMPT INJECTION] (SAFE): The skill processes untrusted data such as branch names and commit messages. It mitigates shell injection risks in Step 4 by using a single-quoted HEREDOC (<<'EOF') when generating PR bodies, which prevents the shell from performing variable expansion or command substitution on the content.
  • Ingestion points: Git branch names, commit messages, and worktree paths (SKILL.md).
  • Boundary markers: Utilizes <<'EOF' to isolate the PR body content from the shell command.
  • Capability inventory: Git repository management, local test suite execution, and GitHub CLI interactions (SKILL.md).
  • Sanitization: Employs literal HEREDOCs to prevent accidental execution of commands embedded in branch metadata or commit descriptions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:22 PM