The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Python's subprocess module in several scripts (scripts/run_eval.py, scripts/improve_description.py, and eval-viewer/generate_review.py) to execute helper scripts and interact with the claude CLI. This functionality is essential for the skill's primary purpose of automating the skill development and optimization workflow.
[EXTERNAL_DOWNLOADS]: The HTML viewer templates (eval-viewer/viewer.html and assets/eval_review.html) reference external JavaScript and CSS resources from well-known and trusted services. Specifically, it uses the SheetJS library from cdn.sheetjs.com for processing Excel files and Google Fonts from fonts.googleapis.com. These are standard UI dependencies and do not represent a security risk.
[DATA_EXFILTRATION]: To provide a user-friendly interface for reviewing test results, eval-viewer/generate_review.py implements a local HTTP server using the standard Python library. It serves data only to the local machine (localhost) and is used solely to facilitate the human-in-the-loop review process.
[REMOTE_CODE_EXECUTION]: While the skill executes local scripts via the shell, all scripts are bundled within the skill itself. It does not download or execute arbitrary code from untrusted remote sources.

skill-creator