The Agent Skills Directory

[SAFE]: Comprehensive architectural references and templates were reviewed for malicious patterns and prompt injection; all content was found to be professional and focused on legitimate AI engineering practices.
[SAFE]: Technical guidance within the skill includes proactive security recommendations, such as DNS rebinding protection for local MCP servers and scoped data isolation for autonomous agents.
[SAFE]: Python utility scripts like pipeline_orchestrator.py and prompt_optimizer.py were analyzed for dangerous function calls. These scripts utilize yaml.safe_load for parsing configuration and perform local analysis without unauthorized network operations.
[SAFE]: The use of compile() in scripts/pipeline_orchestrator.py is specifically for static syntax validation of generated pipeline code. The compiled code is not executed, which is a safe and common pattern for code generation tools.

ai-engineering