The Agent Skills Directory

[SAFE]: The skill is a collection of reference documentation and implementation patterns. No malicious instructions, prompt injections, or obfuscated payloads were detected within the markdown or code snippets.- [COMMAND_EXECUTION]: The skill provides educational code snippets in Go, Python, Node.js, and Rust demonstrating how to invoke CLI tools using standard subprocess libraries. These examples correctly emphasize safety practices, such as using non-interactive flags (--yes), handling timeouts, and separating stdout for data from stderr for logs.- [CREDENTIALS_SAFE]: The documentation outlines secure authentication flows including OAuth 2.0, token-based authentication via stdin/files, and workload identity. Code examples follow best practices by using environment variables or secure inputs rather than hardcoding secrets or passing them via command-line arguments.- [DATA_EXFILTRATION]: There is no evidence of unauthorized data collection or exfiltration. The network patterns described relate to standard API interactions for CLI functionality.

optimize-cli-for-agents