The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands that interact with cryptocurrency exchanges.
Evidence: Commands such as ritmex-bot order create, ritmex-bot order cancel-all, and ritmex-bot strategy run are used to perform high-stakes financial operations.
[EXTERNAL_DOWNLOADS]: The skill utilizes package runners to obtain the necessary tools from remote registries at runtime.
Evidence: The instructions recommend using npx ritmex-bot and bunx ritmex-bot, which download packages from the npm or Bun registries.
[REMOTE_CODE_EXECUTION]: Code fetched from external registries is executed immediately within the agent's environment.
Evidence: The use of npx and bunx leads to the execution of code that is not part of the skill's static distribution.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it parses data originating from external cryptocurrency exchanges.
Ingestion points: JSON output generated by ritmex-bot (sourced from exchange API responses) is parsed by the agent.
Boundary markers: There are no explicit markers or instructions provided to the agent to treat the CLI output as untrusted or to ignore embedded instructions.
Capability inventory: The agent has access to sensitive capabilities, including order creation and strategy execution.
Sanitization: The skill does not define any sanitization or validation logic for the data received from external sources before it is processed by the agent.

use-ritmex-bot