The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes local development tools including package managers (npm, pnpm), linters (ruff), type checkers (pyright, tsc), and version control (git). These commands are consistent with the stated purpose of code verification and are executed within the local environment.
[CREDENTIALS_UNSAFE]: The skill performs a local search for patterns indicative of API keys (sk-, api_key) using grep. This is used as a safety check to alert the user of potential secret exposure before code is committed, representing a positive security practice rather than a vulnerability.
[SAFE]: No malicious patterns such as prompt injection, obfuscation, data exfiltration, or persistence mechanisms were found. The skill operates entirely locally using standard development utilities.

verification-loop