The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data provided by the user, such as wallet addresses, contract identifiers, and transaction amounts.
Ingestion points: User-supplied addresses and transaction parameters are used as inputs for blockchain operations described in SKILL.md.
Boundary markers: The skill includes a dedicated 'Security Rules' section that mandates input validation and explicit user confirmation before any write operations are executed.
Capability inventory: The skill facilitates balance queries, token transfers, and allowance approvals using external libraries like viem and @solana/kit.
Sanitization: It instructs the agent to 'ALWAYS validate all inputs' and 'verify chain ID matches expected environment' to mitigate risks associated with malformed or malicious inputs.
[EXTERNAL_DOWNLOADS]: The skill references and downloads official dependencies and resources.
Evidence: Instructions to install standard packages including viem, @solana/kit, @solana-program/token, ws, dotenv, and bs58.
Evidence: Fetches USDC contract addresses and utilizes faucets from official Circle and Solana domains (developers.circle.com, faucet.circle.com, faucet.solana.com).

use-usdc