依最近已完成的 8 小時觀測回合間的 GitHub 星號增量排序
Dive deep into your data. A fast, native, multi-engine database client written in Rust. Free and open source, no telemetry.
翻譯準備中
Scikit-learn for Recursive Self-improvement (RSI). A modular framework for recursive harness evolution.
翻譯準備中
正在載入完整 README。
檢視完整 README ↗A modular framework for recursive harness evolution.
English | 简体中文 · User guide · Examples
Gear is an open-source optimization framework for improving AI agent performance on real-world tasks.
To tailor an agent to a target scenario, first prepare a benchmark of representative tasks with clear success criteria. You can use an existing benchmark or build your own task set. Run the Refine Skill on that benchmark. Gear uses its evaluation results to iteratively improve the agent’s model, instructions, tools, and workflows.
The two examples below compare Luna Max with a Gear-optimized harness against Astra Max with Codex on the same AutomationBench tasks. Follow their execution from reading source information and applying rules to delivering the final result.
Each group keeps the model and reasoning effort fixed. Luna compares the base DSH harness with Gear's optimized DSH harness; Astra compares Codex with Gear's optimized DSH harness.
| Model (effort) | Harness | Marketing pass rate | Marketing objective completion | Support pass rate | Support objective completion | Terminal-Bench 2.1 pass rate |
|---|---|---|---|---|---|---|
| GPT 5.6 Luna (medium) | DSH (Base) | 27.0% | 75.4% | 17.0% | 72.9% | 52.8% |
| GPT 5.6 Luna (medium) | DSH + Gear | 40.0% | 83.9% | 25.0% | 81.3% | 62.9% |
| Δ (percentage points) | +13.0 | +8.5 | +8.0 | +8.4 | +10.1 | |
| GPT 6 Astra (max) | Codex | 57.0% | 84.1% | 35.0% | 88.4% | 86.7% |
| GPT 6 Astra (max) | DSH + Gear | 61.0% | 86.3% | 39.0% | 88.7% | 87.6% |
| Δ (percentage points) | +4.0 | +2.2 | +4.0 | +0.3 | +0.9 |
The optimized DSH harness was also evaluated with GPT 5.6 Luna at max reasoning effort:
| Benchmark | Task pass rate | Objective completion |
|---|---|---|
| AutomationBench / Marketing | 53.0% | 88.9% |
| AutomationBench / Support | 27.0% | 82.4% |
| Terminal-Bench 2.1 | 84.3% | — |
These are separate evaluations at a higher reasoning effort; improvements over Luna's medium baseline include both harness and effort changes. The dash means the metric is not reported.
On Marketing, Luna max with Gear achieved 88.88% objective completion, compared with 84.08% for Astra max in Codex; their task pass rates were 53% and 57%, respectively. In a separate cost comparison with both models using the same optimized DSH harness (a0740800), Luna max's API cost averaged $0.0568 per task, versus $1.1858 for Astra max — 95.21% lower. See the token usage and cost methodology.
Use Gear through its Refine Skill in Codex, Claude Code, DSH, or another compatible agent environment. Start with an existing benchmark or provide your own tasks in Harbor format.
For installation and dependency setup, see the quick start guide.
Let your agent install Gear. Paste the following prompt into your agent session:
Follow https://rsigear.xyz/docs/gear/quickstart to install Gear in my environment.
Install rsi-gear@latest and agent-hitch@latest using npm and verify that all required dependencies are available.
Install the complete Refine Skill bundled with Gear in my current agent environment and configure its connection to Gear.
Use my actual task paths, target harness, and model settings; ask me for any missing information.
Verify that the Refine Skill can connect to Gear, then report the verification result and explain how to start my first optimization run.
For manual installation, install Gear and Hitch, the evaluation runner. The following example also installs DSH as the task agent runtime:
npm install --global rsi-gear@latest agent-hitch@latest @deepseek-ai/dsh@latest
hitch eval setup harbor
Follow the setup guide to install the bundled Refine Skill in your agent environment and configure your benchmark. Then invoke /refine where supported, or submit the following request:
Use the Refine Skill to improve performance on AutomationBench's Marketing tasks.
Use Codex + Astra to propose improvements and DSH + Luna to run the tasks.
Run one round of optimization.
The agent that proposes changes is called the Meta agent. It can use a different model from the task agent that executes the benchmark tasks.
以上為 README 的開頭部分。其餘內容請閱讀原文。 檢視完整 README ↗