The Agent Skills Directory

[SAFE]: The skill is strictly advisory. It contains explicit instructions for the AI to never execute tasks, write code, or create files, ensuring it remains within its intended scope as a prompt optimizer.
[SAFE]: Project detection is performed by reading standard, non-sensitive manifest files (such as package.json, go.mod, and pyproject.toml) to determine the technical stack. This is a legitimate function for a development assistant and involves no data exfiltration.
[SAFE]: No malicious patterns such as obfuscation, hidden URLs, remote code execution, or credential harvesting were found. The skill follows best practices by encouraging structured workflows and human-in-the-loop verification.

prompt-optimizer