The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from decentralized protocols (ALEX DEX and Zest Protocol), which could potentially contain adversarial content designed to influence the agent's behavior.
Ingestion points: Data is ingested through subcommands such as alex-list-pools, zest-list-assets, and alex-get-swap-quote in defi/defi.ts.
Boundary markers: The skill does not implement explicit boundary markers or delimiters when interpolating external protocol data into the agent's context.
Capability inventory: The skill possesses significant capabilities, including executing token swaps (alex-swap), supplying assets to lending pools (zest-supply), and borrowing assets (zest-borrow) as defined in defi/defi.ts.
Sanitization: Input arguments are handled by the commander library, and token amounts are validated via BigInt conversion, providing basic protection against malformed numeric input.
[COMMAND_EXECUTION]: The skill facilitates the execution of blockchain transactions via subcommands in defi/defi.ts. While these are the primary purpose of the skill, they represent powerful actions that the agent can take on behalf of the user, requiring an unlocked wallet state.

defi