2026-11-06 –, Lotus Stage 3
LLMs wake up with a stale picture of Ethereum. eth-evals measures it (242 tasks, 14 models, 58–82%). ethskills.com patches it. Then I gave the skill to my agent, clawdbotatg, and let it write, audit and deploy 14 dapps and a ~$67K staking contract on Base with no human reading the code. Most of this talk is the post-mortem: it leaked its own key twice, the audit pipeline rubber-stamped things it shouldn't have, and here's what I changed.
I built eth-evals to measure how much LLMs actually know about Ethereum: 242 closed-book tasks, 25 categories, deterministic grading. Fourteen models score 58–82%, and they all fail in the same places. ethskills.com is the patch: 24 skill files. Load the skill and Opus jumps from 33% to 95% (on tasks keyed to the skill — a ceiling; held-out numbers by Devcon).
Then I handed the skill to my agent, clawdbotatg. It writes Solidity, audits itself with its own $1 audit service (x402 payments, ERC-8004 reviews), and deploys with no human review: 14 dapps on Base, 714 paid jobs, ~690 audits, a staking contract holding about $67K. It also leaked its own private key. Twice. We'll dig into where the audits worked, where they didn't, and how I locked the agent down after.
The same agent co-hosts slop.computer with me: 50 interviews with crypto × AI builders on a multisig with Kohaku private txs and Interfold private votes. I'll share what keeps coming up.
All of it is open source and live.
Austin Griffith. Ethereum Foundation; founder of BuidlGuidl and Scaffold-ETH. Built ethskills.com, eth-evals, and clawdbotatg, an AI agent that writes, audits and deploys its own contracts and co-hosts slop.computer.