The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data from the OKX API via CLI tool outputs, creating a surface for indirect prompt injection. \n- Ingestion points: Financial data and account states enter the context through JSON outputs of okx CLI commands as defined in SKILL.md, savings-commands.md, and dcd-commands.md. \n- Boundary markers: The skill requires the agent to format all outputs as Markdown tables and necessitates explicit user confirmation before executing any commands that modify account states or move funds. \n- Capability inventory: The skill can execute authenticated CLI commands for reading account states and performing financial transactions (e.g., purchase, redeem) as documented in onchain-commands.md and autoearn-commands.md. \n- Sanitization: No specific technical sanitization or validation of the retrieved API data is described before the agent processes it for display or logic. \n- [EXTERNAL_DOWNLOADS]: The skill frontmatter and installation instructions describe fetching the @okx_ai/okx-trade-cli package from npm. This tool is a vendor-provided resource designed for interacting with OKX services.

okx-cex-earn