Agents / Skills
Three-Agent Harness
A multi-agent prompt harness for autonomous app development without self-grading.
Is this your creation?
Description
An implementation of Anthropic's multi-agent harness design that orchestrates three specialized agents—Planner, Generator, and Evaluator—around a single prompt. It features a Playwright-based evaluation loop that prevents agents from grading their own work, allowing for long-running autonomous development sprints that produce playable apps and games.
Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.