Skip to main content

Quick Answers


Do I need to know how to code?

No. You need to be able to describe what you want. If you can explain an app idea to a friend, you can build it with Claude and the buidl-opnet-plugin.


What's OPNet and why does it matter?

OPNet is smart contracts on Bitcoin L1. Not a sidechain, not a wrapped version — actual Bitcoin. Your contracts are secured by Bitcoin's proof-of-work. Your users pay with real BTC. It's the most secure platform you can build on.


What's the buidl-opnet-plugin?

It's a Claude Code plugin that transforms Claude from a generic coding assistant into an OPNet specialist. It loads 8 specialized agents (contract dev, frontend dev, backend dev, auditor, deployer, testers), OPNet-specific knowledge slices, and the /buidl command that runs the full build pipeline automatically.


What's the difference between claudey and claudeyproj?

claudey is safe mode — Claude asks for approval before running commands. Good for learning.

claudeyproj is autonomous mode — Claude runs the whole pipeline without stopping. This is what you use with /buidl. The pipeline is designed to run continuously; interrupting it slows everything down.


What does /buidl actually do?

It runs a full development pipeline: your idea gets clarified, a spec gets written, the smart contract gets built, the frontend and backend get built in parallel, an auditor reviews everything against 27 known bug patterns, the contract gets deployed, and tests run. All automatically.

For a medium-complexity app, this takes 15-30 minutes and produces working, audited code.


Do I need to explain OPNet in my prompts?

No. The plugin loads OPNet knowledge into each agent before they start. The contract dev agent knows OPNet token standards, SafeMath requirements, storage patterns, and deployment flows. You don't need to explain any of this. Focus your prompts on what you want, not how to build it.


What's the MCP for? Do I need it if I have the plugin?

Both serve different purposes. The buidl-opnet-plugin gives Claude the multi-agent build pipeline and OPNet development knowledge. The opnet-bob MCP gives Claude live access to blockchain data — current token prices, contract state, transaction status. Add both. They complement each other.


How much does this cost?

Claude Code requires an Anthropic account. The Pro plan ($20/month) is recommended for running /buidl pipelines — they do significant work. The plugin is free. The MCP is free. OPWallet is free. Deploying contracts costs a small amount of BTC (for the Bitcoin transaction fees).


Can I deploy to mainnet?

Yes. The deployer agent handles both testnet and mainnet. Always test on testnet first. When you're ready for mainnet, tell Claude to deploy to mainnet and make sure your OPWallet has real BTC.


What if the /buidl pipeline gets interrupted?

Run /buidl-resume. The pipeline uses atomic state writes and checkpointing — if it gets interrupted (network issue, terminal closed, timeout), it can pick up where it left off instead of starting over.


What's a .btc domain?

A .btc domain is a Bitcoin-native web address. Your app lives at yourname.btc instead of a random URL. It's decentralized, you own it, and it's resolved via OPNet infrastructure. Set it up with the opnet-cli: opnet deploy your-domain ./dist.


My app doesn't look very good. How do I fix it?

Ask for it explicitly. Tell Claude: "Make this look like a real DeFi app. Dark background, OPNet orange accents (#F7931A), glassmorphism cards, smooth transitions. Production-ready, not a tutorial demo." Be specific about what looks off and what you want instead.


What's the auditor agent doing exactly?

The opnet-auditor agent checks your code against 27 known bug patterns — 9 marked CRITICAL — pulled from real vulnerabilities found in production OPNet repos. Things like storage pointer collisions, u256 overflow paths, missing CSV timelocks, incorrect UTXO handling. It catches these before you deploy.


Can I build on mainnet without testing on testnet?

Technically yes. Practically, don't. Testnet deployments cost nothing and catch deployment-specific bugs. Always test the full flow on testnet first: deploy, connect wallet, do the main thing, verify it works. Then deploy to mainnet.


The pipeline built something but it's not quite what I wanted. Do I run /buidl again?

No. Use direct prompts to iterate: "Change X to Y" or "Add Z feature." The /buidl pipeline is for initial builds. For changes after the build, just talk to Claude directly. Running /buidl again would start over from scratch.


Where do I get help if I'm stuck?

The OPNet Discord is the best place. Post what you're building, what's happening, and what you've tried. The community builds on OPNet, they've hit the same walls you're hitting, and they'll help.


Is vibecoding "real" development?

The code Claude writes is real code. The contracts that get deployed are real contracts running on real Bitcoin L1. The apps users interact with are real apps. The process of getting there is different from typing every line yourself, but the output is the same.

Call it whatever you want. The apps are real.

Ready to test your knowledge?

20 questions covering everything from vibecoding basics to shipping a complete dApp.

Take the Quiz