The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill executes common Ruby development tools (bundle exec standardrb, bundle exec erblint, bin/brakeman). These are standard operations for the skill's primary purpose of code quality and security scanning.
[REMOTE_CODE_EXECUTION] (SAFE): No remote code patterns or untrusted downloads were detected. The skill relies on local environment dependencies.
[DATA_EXFILTRATION] (SAFE): There are no network operations or attempts to access sensitive system files like credentials or SSH keys.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill is designed to process external Ruby and ERB files which could contain malicious content.
Ingestion points: Local Ruby (.rb) and ERB (.erb) files.
Boundary markers: None explicitly defined in instructions.
Capability inventory: Execution of shell commands via bundle exec and local binary execution of bin/brakeman.
Sanitization: Relies on the inherent static analysis logic of the linter tools to safely parse input files.

lint