AgentLoop

Starts a fresh Codex worker and critic every cycle

Data & analytics Product Hunt Launched Jul 23, 2026 by Edward Yi aiedwardyi View original post ↗

Unlike long-running agent chats, AgentLoop starts a fresh Codex worker and critic every cycle. Set the goal and GUIDELINES.md rubric once; workers build, critics test, and failures become concrete fix notes for the next clean context. Project files carry the memory. Runs stay local, sandboxed, observable, and cancellable from a live dashboard, with ChatGPT control through MCP. Polish mode can continue beyond PASS until the critic says SHIP. Open source and zero-dependency Node.js.

Open SourceDeveloper ToolsArtificial IntelligenceGitHubOpenAI Day

More Data & analytics

CheckBodyFat DataPulse TableTap trustyourinbox Data Lens