The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructions direct the agent to download and execute a shell script from a remote URL (https://raw.githubusercontent.com/6551Team/openskills/main/skills/opentrade/install.sh) using curl | sh if the tool is not found, needs updating, or encounters an error. This pattern allows the execution of code hosted on a remote server.\n- [COMMAND_EXECUTION]: The skill performs multiple local command executions, including which for dependency verification, date and cat for managing update intervals, and several calls to the opentrade CLI tool to perform quote and swap operations.\n- [EXTERNAL_DOWNLOADS]: The skill fetches installation scripts and version-check configuration from the developer's GitHub repository (github.com/6551Team).\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it processes the output of external CLI commands to determine transaction parameters.\n
Ingestion points: JSON responses from opentrade trade routers and opentrade swap quote.\n
Boundary markers: The skill expects specific JSON schemas but does not define explicit sanitization for the content of those fields.\n
Capability inventory: The skill can generate and approve blockchain transaction data, which is then broadcast to the network.\n
Sanitization: No explicit sanitization or validation of the CLI tool's output is mentioned before using the data in subsequent commands.

opentrade-dex-swap