The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches market and user data from Hyperliquid's public API endpoints, integrating external information into the agent's workspace.\n
Ingestion points: hyperliquid_client.py retrieves data from api.hyperliquid.xyz via POST requests to the /info endpoint.\n
Boundary markers: The script produces structured output (tables or JSON) without specific delimiters for prompt isolation.\n
Capability inventory: The skill can perform network reads (urllib.request), output data to the terminal, and write JSON files to the local disk (export command).\n
Sanitization: Data is parsed using the standard json library and specific fields are selected for display.\n- [SAFE]: The skill implements standard configuration practices by checking environment variables and local .env files for the API URL and user address. It does not require or manage sensitive credentials such as private keys.

hyperliquid