LMMs Lab open-sources The Harness, an MIT-licensed autonomous agent framework
liuziwei7 · x · 2026-11-14
LMMs Lab open-sourced The Harness, described as 'Automatic Control for intelligence in action'—an agent framework that runs on its own, thinks while it acts, and learns as it works. Released under the MIT license for free commercial use. The repo is live with 120+ commits, docs, tests, and a Claude plugin; the team notes it's early-stage and welcomes feedback.
More from coding & agent
- Keyfleet to open fleet creation with group access rules and personal persistent-agent teams — seanwbren · 2026-11-14
- Regents Labs launches 6 agent-native apps with token-owned revenue split in USDC on Base — seanwbren · 2026-11-14
- Tokyo Now: a pixel-art live map of Tokyo refreshed every 5 minutes by an AI model — jasonkneen · 2026-11-14
- InferenceForge claims 43% higher LLM inference throughput than best AgentX baseline — kaiwei_chang · 2026-11-14
- Anthropic ships Self-Hosted Claude Managed Agents with first-party Evals product — threepointone · 2026-11-14
- David Khourshid: agents will force a rethink of libraries built to save developers typing — DavidKPiano · 2026-11-14