The Agent Skills Directory

[SAFE]: The skill implements proactive security measures by explicitly instructing the agent never to read the sensitive credential file ~/.volcengine/config.json and to ensure that access keys or secret keys are never exposed in the output.
[COMMAND_EXECUTION]: The skill manages cloud resources via the ve command line tool. It defines a mandatory safety framework in SKILL.md that classifies operations into Read-only, Write, and Destructive, requiring explicit user approval for any action that modifies or deletes resources.
[EXTERNAL_DOWNLOADS]: The helper scripts scripts/fetch_swagger.py and scripts/find_api.py perform network requests to api.volcengine.com to retrieve API documentation and search results. These operations target official vendor infrastructure and are used to provide the agent with accurate parameter definitions.

volcengine-cli