skills/tobihagemann/turbo/finalize/Gen Agent Trust Hub

finalize

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Orchestrates version control tasks using Git (git add, git diff, git status, git push) and GitHub CLI (gh pr view) to manage the repository state.
  • [COMMAND_EXECUTION]: Invokes project-specific test suites and linters/formatters, which involves executing code and scripts defined within the local project environment.
  • [PROMPT_INJECTION]: Vulnerable to indirect prompt injection because the skill processes untrusted code and test outputs from the repository.
  • Ingestion points: Reads file content and diffs via git diff, grep, and file system searches in Phase 1 and Phase 3.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious directions embedded in the source code it processes.
  • Capability inventory: Possesses capabilities to modify the local file system (git add), execute shell commands (tests/lints), and push changes to remote repositories (git push).
  • Sanitization: Implements a specific safety rule to identify and warn against staging sensitive files like .env, credentials, or API keys, which mitigates accidental data exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 04:06 PM