The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the @3rd-eye-labs/openmm package from the npm registry during the installation phase. This package originates from a source that is neither the skill author (qbtlabs) nor a recognized trusted organization.
[CREDENTIALS_UNSAFE]: The skill requires multiple highly sensitive API keys and secrets for financial exchanges (MEXC, Gate.io, Kraken, Bitget) to be provided via environment variables. These credentials provide the agent with direct access to trade and manage funds on these platforms.
[COMMAND_EXECUTION]: The skill utilizes the Bash(openmm:*) capability to execute trading commands. This allows the AI agent to place, list, and cancel orders. While documentation includes safety guidelines such as user confirmation, the underlying binary is executed with the provided API credentials.
[PROMPT_INJECTION]: A surface for indirect prompt injection exists because the skill processes data from external exchange APIs (order IDs, ticker symbols, balances).
Ingestion points: Data retrieved from exchange APIs via list_orders and get_balance tools.
Boundary markers: None explicitly defined in the provided instruction files.
Capability inventory: Subprocess execution via openmm CLI for trading and account management.
Sanitization: No explicit sanitization or validation of API responses is documented in the skill instructions.

openmm-order-management