Skip to main content
Agent Skills package Whitechain-specific instructions and reference files that an AI agent loads on demand, rather than relying on the model’s general OP Stack knowledge. The whitechain-dev skill is a testnet developer playbook. It is not tied to one vendor’s agent: any agent that supports the Agent Skills format, including Claude Code, can load it. It assumes you control your own signing key, for example a Foundry keystore or an environment variable, and covers network setup, contract deployment, verification, and chain inspection on Whitechain Sepolia. The skill does not cover token transfers, swaps, signing messages, or bridging. Support for those end-user wallet actions is planned separately.

​
What it covers

​
Safety guardrails

The skill never accepts a private key pasted in chat; it expects a Foundry encrypted keystore or an environment variable. Before any write operation, such as a contract call or deployment, it shows a summary and waits for explicit confirmation. It validates the chain ID before signing, to confirm the target is Whitechain Sepolia (1874), and validates shell inputs before building forge, cast, or solc commands.

​
Install and use

  1. Install the skill:
  1. Ask your agent to perform a Whitechain task in plain language, for example: “Deploy an ERC-20 token to Whitechain Sepolia and verify it.” A compatible agent, such as Claude Code, loads the skill automatically when it recognizes the task.