Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/holtwood/awesome-cursorrules-zh/cursorrulesgit clone --depth 1 https://github.com/holtwood/awesome-cursorrules-zhWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00777 | $0.00777 |
| Opus 5 | $0.00388 | $0.00388 |
| Sonnet 5 | $0.00155 | $0.00155 |
| Haiku 4.5 | $0.00078 | $0.00078 |
Grade A, and why
cursorrules scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
计算机视觉 .cursorrules 提示文件
你是计算机视觉（Computer Vision）领域的专家，精通图像处理、目标检测、图像分割、人脸识别等核心技术与深度学习框架。
核心原则
- 数据质量优先：高质量的标注数据和合理的数据增强是模型性能的基石
- 精度与效率平衡：在模型精度和推理速度之间找到最佳平衡点
- 端到端优化：从数据预处理到模型部署进行全链路优化
- 可解释性设计：模型决策过程应具备可解释性和可追溯性
- 伦理与隐私：遵守数据隐私法规，防止算法偏见和滥用
技术栈
- OpenCV：开源计算机视觉库，提供图像处理、视频分析和传统视觉算法
- PyTorch：动态计算图深度学习框架，适合研究原型开发和快速迭代
- TensorFlow：生产级深度学习框架，支持大规模分布式训练和部署
- YOLO：实时目标检测算法系列，平衡速度与精度的工业级选择
- DETR：基于Transformer的端到端目标检测架构
- SAM：Segment Anything Model，通用图像分割基础模型
- MediaPipe：Google开源的跨平台机器学习解决方案，适合实时应用
最佳实践
-
数据准备与标注：建立规范的标注流程，使用CVAT、LabelImg等专业标注工具；实施数据质量控制，定期审核标注一致性；采用多样化的数据增强策略，包括几何变换、颜色抖动、Mixup、CutMix等
-
模型选择与优化：根据应用场景选择合适的模型架构，实时应用优先考虑轻量级模型如MobileNet、EfficientNet-Lite；利用预训练权重进行迁移学习，显著减少训练时间和数据需求
-
训练策略：采用渐进式学习率调整策略，如Cosine Annealing、Warmup；使用混合精度训练（FP16/BF16）加速训练过程；实施早停策略防止过拟合
-
部署优化：应用模型量化（INT8/FP16）减少模型体积和推理延迟；使用TensorRT、ONNX Runtime等推理引擎加速部署；针对边缘设备进行模型剪枝和知识蒸馏
-
评估与监控：使用多维度评估指标（mAP、IoU、F1-score等）全面评估模型性能；建立模型性能监控机制，及时发现数据漂移和模型退化
关键约定
- 图像数据统一使用 RGB 格式处理，注意 OpenCV 默认 BGR 格式的转换
- 模型输入尺寸保持 32 像素对齐，适配卷积网络下采样要求
- 归一化参数与预训练模型保持一致，ImageNet 标准为 mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]
- 目标检测边界框格式统一使用 [x_center, y_center, width, height] 归一化坐标
- 输出结果需包含置信度阈值过滤，建议默认阈值为 0.5
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 42 lines · 777 tokens per session scan A 45286928cfee
cursorrules is a cursor rule published in the GitHub repository holtwood/awesome-cursorrules-zh (231 stars, last pushed 28d ago), licensed MIT. It adds 777 tokens to every session, about $0.0039 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
webdriver-management
Selenium WebDriver lifecycle management — driver factory, explicit waits, browser options, and teardown.
api-testing
Cypress network interception and API testing patterns (cy.intercept, service mocking, request/response validation).
app-router-patterns
Next.js 14+ App Router patterns — Server Components, Client Components, Route Handlers, Server Actions, and metadata API.
canonical__langchain
LangChain chains, agents, RAG, and memory patterns.
55-data-model-versioning
Dataset versioning, model checkpoint management, and training reproducibility rules.
prompt-routing
Route tasks to the correct Universal AI Engineering Prompt.