The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill's installation manifest and the setup.sh script download a pre-built binary (gate-dex) from the vendor's official domain at https://gate-dex-cli.gateweb3.cc. This is a standard installation procedure for this skill ecosystem.
[COMMAND_EXECUTION]: The skill uses the gate-dex CLI to perform various blockchain queries and allows for arbitrary JSON-RPC calls through the rpc subcommand. While the skill is described as read-only, the tool executes commands directly in the shell environment.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it takes untrusted user input, such as token addresses and RPC parameters, and uses them as arguments in shell commands.
Ingestion points: User-provided token addresses, symbols, and JSON-RPC method parameters specified in SKILL.md and references/scenarios.md.
Boundary markers: There are no explicit boundary markers or instructions to the AI to sanitize or escape user input before interpolation into CLI arguments.
Capability inventory: The skill can execute various subcommands of the gate-dex binary and perform network requests to blockchain RPC endpoints.
Sanitization: The skill relies on natural language instructions for the AI to ensure parameters are valid JSON, rather than providing programmatic sanitization for shell execution.

gate-dex-market-cli