The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the gmgn-cli@1.0.1 package from the npm registry. This is a vendor-specific tool provided by the skill author and is considered a trusted resource in the context of the skill's functionality.
[COMMAND_EXECUTION]: The skill executes shell commands using the gmgn-cli binary to fetch token information, security metrics, and holder data. These commands incorporate user-supplied and API-retrieved arguments.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface where data retrieved from external API responses (token addresses) is interpolated into CLI commands.
Ingestion points: Token addresses are retrieved from external API responses and passed as arguments to the CLI.
Boundary markers: The documentation includes notes to validate that addresses match expected chain formats (e.g., base58 for Solana, hex for BSC/Base) to mitigate malicious input.
Capability inventory: Execution of subprocesses via the gmgn-cli tool which interacts with the local system and network.
Sanitization: The CLI tool is documented to enforce input validation at runtime, though the reliance on external API data for command arguments remains a structural risk.

gmgn-token