The Agent Skills Directory

[SAFE]: The skill operates entirely locally using provided markdown references and Python scripts. No network requests, external downloads, or sensitive file accesses are performed.
[COMMAND_EXECUTION]: The Python scripts (framework_analyzer.py and prompt_evaluator.py) perform safe string analysis and logic without invoking shell commands or dynamic execution (eval/exec) on user-provided data.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user prompts as its primary function. While it generates revised instructions based on this input, it does not execute them automatically. It provides clear copy-paste instructions for the user, maintaining a human-in-the-loop safety boundary. This is a standard functional surface for prompt engineering tools.
[PROMPT_INJECTION]: No attempts to override system safety guidelines or bypass AI constraints were found in the instructional content.

prompt-architect