The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to install the official @volcengine/cli package via npm and references the vendor's GitHub repository for manual binary downloads.
[COMMAND_EXECUTION]: The skill facilitates cloud infrastructure management via the ve command. It enforces safety protocols that require the agent to obtain user confirmation for any write or destructive operations (e.g., Create, Run, Delete, Stop) and suggests using dry-run validation where available.
[PROMPT_INJECTION]: The skill processes data from the Volcengine API at runtime, creating a potential surface for indirect prompt injection.
Ingestion points: Documentation and search results are dynamically retrieved from https://api.volcengine.com using the scripts/fetch_swagger.py and scripts/find_api.py scripts.
Boundary markers: The skill does not define explicit delimiters or 'ignore' instructions when presenting the fetched API metadata to the agent, relying on the agent to parse natural language descriptions for parameter information.
Capability inventory: The agent has access to Bash, Read, and Write tools, enabling it to execute commands or modify local files based on the fetched data.
Sanitization: The Python scripts perform basic markdown escaping (e.g., pipe character replacement) and content truncation, but do not sanitize the semantic content of the API descriptions.

volcengine-cli