Harness — page 1

• Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams
  arxiv:2606.01770 · 2026-06-01 · text available
• Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
  arxiv:2604.25850 · 2026-04-28 · text available
• DeepSeek Harness
  url:github.com/deepseek-ai/deepseek-harness · 2026-08-13 · metadata only
• Duck Harness: Winning Solution for ARC-AGI-3 Milestone 1 (Tufa Labs)
  url:tufalabs.ai/research/duck-harness · 2026-07-01 · metadata only
• Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering
  arxiv:2604.08224 · 2026-04-09 · text available
• First evidence of recursive self-improvement (AIDE2, Weco)
  url:weco.ai/blog/first-evidence-of-recursive-self-improvement · 2026-07-14 · metadata only
• Harness Engineering (OpenAI)
  url:openai.com/index/harness-engineering · 2026-08-25 · metadata only
• HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?
  arxiv:2609.01437 · 2026-09-01 · text available
• Harnesses for models (Lilian Weng, 2026)
  url:lilianweng.github.io/posts/2026-07-04-harness · 2026-07-04 · metadata only
• Language model harnesses are compositional generalizers (Alex Zhang)
  url:alexzhang13.github.io/blog/2026/harness · 2026-07-20 · metadata only
• Meta-Harness: End-to-End Optimization of Model Harnesses
  arxiv:2603.28052 · 2026-03-30 · text available
• Prime Agent: a self-improving RLM agent (Prime Intellect)
  url:github.com/primeintellect-ai/prime-agent · 2026-05-08 · metadata only
• PRO-LONG: Programmatic Memory Enables Long-Horizon Reasoning
  url:github.com/alexisfox7/rgb-agent · 2026-07-22 · metadata only
• Schema: a symbolic world model for ARC-AGI-3
  url:schema-harness.github.io · 2026-07-17 · metadata only
• Structured Scaling of AI Discovery Across Diverse Scientific Domains
  arxiv:2604.19341 · 2026-04-21 · text available