reharness
Self-improving LLM agent harness, bounded recursive self-improvement: a Revision Session reads the agent's own returned run, rewrites its tools, hooks, skills and prompts, and proves the change with a Live Probe on the same request. Per-task adoption, no answers shown. Measured on AutomationBench.
- agent-harness
- agentic-ai
- ai-agent
- ai-agents
- anthropic
- automationbench
- autonomous-agents
- claude
- continual-learning
- harness
- harness-engineering
- llm
- llm-agents
- python
- recursive-self-improvement
- Stars
- 6
- Forks
- 2
- + today
- +1
- Created
- 1mo
Ranking data as of October 28, 2026 (UTC).
Star History
Today, hour by hour
Overview
Self-improving LLM agent harness, bounded recursive self-improvement: a Revision Session reads the agent's own returned run, rewrites its tools, hooks, skills and prompts, and proves the change with a Live Probe on the same request. Per-task adoption, no answers shown. Measured on AutomationBench. It ranks #1375 on GitTiger, gaining +1 star on October 28, 2026 (UTC).
The project is written in Python and has 2 forks. It was created 1mo ago.
git clone https://github.com/e13ven-arch/reharness.git
cd reharness
# see README for setup