The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructions direct the agent to run okx config show --json which reads the ~/.okx/config.toml file. This file contains API keys and other sensitive authentication data for the OKX exchange. The agent uses this information to verify the user's trading profiles and environment.
[EXTERNAL_DOWNLOADS]: The skill relies on the installation of the @okx_ai/okx-trade-cli package from the npm registry. This is an official CLI tool provided by the vendor (OKX) for managing trading operations.
[COMMAND_EXECUTION]: The skill executes multiple shell commands through the okx CLI to perform operations such as okx bot grid create, okx bot grid stop, and okx bot dca amend. These commands directly manage financial assets and bot configurations.
[PROMPT_INJECTION]: An indirect prompt injection surface exists as the skill processes data from external tool outputs.
Ingestion points: The agent parses the JSON output of commands such as okx config show, okx auth status, and okx bot grid orders to make logic decisions.
Boundary markers: There are no specified delimiters or instructions to ignore potential injection patterns within the tool's output.
Capability inventory: The skill possesses the capability to execute financial transactions and modify trading strategies.
Sanitization: The instructions do not include steps for sanitizing or validating the data returned by the CLI before it is used to formulate subsequent commands.

okx-cex-bot