The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill's logic is transparent and focuses on automating a development workflow.
[COMMAND_EXECUTION]: The skill uses local Python scripts (check_release_env.py and scan_skill_release.py) to verify the execution environment and audit target files. These scripts use subprocess calls to interact with standard development tools such as git, node, and npx. All commands are hardcoded or derived from the local environment and do not execute untrusted external input.
[DATA_EXFILTRATION]: While the skill facilitates the publication of data to GitHub, it is designed for user-initiated open-sourcing of their own skills. It includes a robust safety scanner that explicitly blocks the inclusion of API keys, tokens, private keys, and absolute system paths to prevent accidental data exposure.

skill-open-sourcer