The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the okx CLI binary to perform all trading and query operations. Documentation in SKILL.md and reference files (e.g., spot-commands.md, swap-commands.md) outlines how the agent executes these commands to manage orders, positions, and leverage. The skill includes safeguards such as requiring user confirmation before executing write operations and verifying trading modes (live vs. demo).
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the @okx_ai/okx-trade-cli package from the NPM registry. This package is a vendor-owned resource provided by the author 'okx' for interacting with their exchange services.
[INDIRECT_PROMPT_INJECTION]: The skill processes data returned from the OKX API, which represents a potential attack surface for indirect prompt injection if malicious data were present in API responses.
Ingestion points: Data enters the context through read commands such as okx spot orders, okx swap positions, and okx event browse, which fetch order statuses and market information from the exchange.
Boundary markers: The skill instructs the agent to present tool results using user-facing labels in Markdown tables and to maintain a clear distinction between the agent's reasoning and the data retrieved.
Capability inventory: The skill has broad capabilities including placing and amending orders, closing positions, and modifying account settings like leverage across spot and derivative markets.
Sanitization: The skill mandates that the agent confirm key parameters with the user before executing any write commands. Additionally, it explicitly forbids the agent from automatically performing destructive actions suggested by API error messages, requiring the agent to present findings to the user and wait for explicit confirmation instead.

okx-cex-trade