verification-cockpit

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill generates a tmux-layout.sh script that automates the creation of tmux panes. These panes are configured to execute various shell commands discovered within the target project, such as pnpm test, cargo test, and lazygit.
  • [REMOTE_CODE_EXECUTION]: The generated layout script sources external bash libraries located in the user's home directory (~/.config/shell/bin/lib/verify-launch.sh and verify-layout.sh). This behavior establishes a dependency on the local execution environment's shell configuration.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting data from repository files (such as AGENTS.md, package.json, and README.md) to determine which commands to execute. It lacks explicit boundary markers or sanitization for these extracted commands, relying on the user's environment and workflow for safety.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 08:47 AM
Security Audit — agent-trust-hub — verification-cockpit