The Agent Skills Directory

[DATA_EXPOSURE_AND_EXFILTRATION]: The skill interacts with official Hyperliquid API endpoints to retrieve account data such as positions, balances, and orders. It uses an environment variable (HYPERLIQUID_ACCOUNT_ADDRESS) to identify the account, which is a safe practice for managing identifiers. No sensitive credentials like private keys or API secrets are accessed or transmitted.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from the Hyperliquid API which is then parsed and displayed. Ingestion points: Hyperliquid API responses in SKILL.md. Boundary markers: Absent. Capability inventory: curl, jq, and Python SDK. Sanitization: Relies on standard JSON parsing. The risk is safe as the skill's purpose is read-only data display.
[EXTERNAL_DOWNLOADS]: The skill references the hyperliquid Python SDK, which is the official library for the Hyperliquid exchange. No suspicious third-party packages or scripts are downloaded.

hyperliquid-account