跳转到主要内容
alibaba
GitHub 创作者资料

alibaba

按仓库查看 7 个 GitHub 仓库中的 43 个已收集 skills。

已收集 skills
43
仓库
7
更新
2026-08-07
仓库浏览

仓库与代表性 skills

add-new-op
其他计算机职业计算机程序员

为 MNN 框架新增算子。包含 Schema 定义、形状计算、几何计算、后端实现、单元测试的完整 TDD 流程。分 5 步执行,每步有独立测试标准。

2026-03-10
arm-cpu-optimize
软件开发工程师

MNN ARM CPU 算子和低 bit kernel 性能优化。重点覆盖正确性基线、C++ 标量 oracle、C++ SIMD 模拟、寄存器规划、ARM 汇编实现、dispatch/pack 集成、模型级回归和 roofline 性能分析。

2026-07-15
bugfix
网页与数字界面设计师

MNN 各类正确性/回归 bug 的排查方法论集合。按 bug 分类组织:当前覆盖 (1) 内存别名/生命周期错误(buffer aliasing、arena reuse、`MemChunk` / tensor buffer 生命周期)、(2) 量化误差/导出侧权重损坏(低 bit 打包、导出分块、PyTorch MPS/CUDA 大张量静默错误),后续会补充数值精度、并发竞争、图优化回归、Codegen/Shader 错等类别。

2026-08-04
hexagon-optimization
网络与计算机系统管理员

Optimize, refactor, build, and regression-test MNN Hexagon/HVX/HMX DSP code paths under `source/backend/hexagon`, using device profile data, correctness checks, and cDSP crash diagnostics.

2026-08-06
metal-optimize
软件开发工程师

MNN Metal 后端 op/kernel 优化与扩展入口。索引四份 sub-doc:kernel 修改方法论(新加 op / quant bit / dispatcher / weight pack)、LLM 优化知识库(kernel 层 GEMM/GEMV/attention + 系统架构层 + 失败实验档案 + 测量基建)、构建-测试-性能基线、env 开关注册表。根据当前任务选择性阅读对应 sub-doc。

2026-08-05
opencl-optimize
其他计算机职业计算机程序员

MNN OpenCL 后端 kernel 性能优化与新特性集成。覆盖 benchmark 基线、kernel 优化迭代、集成验证全流程,以及新 OpenCL 特性适配、.cl + codegen 双轨、kernel 选路、packed weight 设计、tune 机制、Android 真机验证等参考知识。

2026-07-30
qnn-debug
计算机系统分析师

MNN QNN(高通 HTP/NPU)后端的问题定位、修复与算子适配。覆盖运行时报错(1002/6000/1003/6004)、模型转换/算子校验失败、推理结果异常、精度不达标、以及性能问题;并指导为 QNN 新增/适配算子。核心方法:用 MNN2QNNModel --dump_intermediate_outputs 一次性导出 QNN 全部中间张量,与 CPU(fwd=0) 基线逐张量对比,直接定位首个出错算子(旧的 testMNNFromOnnx.py 截断二分已降为回退手段);区分“真 bug/量化/HTP fp16 精度”;新增算子先查 SDK 算子文档(MasterOpDef/HtpOpDefSupplement)再实现。

2026-08-04
retrospective
其他计算机职业

任务完成后的反思与经验沉淀。仅在任务非平凡且出现调试、失败修复、反复试错、明显误判或可复用教训时使用;简单执行、查询、常规 CI 通过等无新增经验的任务可跳过。

2026-06-30
riscv-cpu-optimize
数据科学家计算机程序员

MNN RISC-V CPU 算子与 LLM 性能优化工作流,覆盖标准 RVV、厂商矩阵扩展(如 SpacemiT IME/IME2)、低比特 GEMM/GEMV/Linear/MatMul、Attention、KV Cache、数据重排、TCM/local-memory 流水和远端开发板验证。用户要求优化或评审 RISC-V/RVV kernel、prefill/decode 性能、量化计算、架构分层、远端板端 benchmark,或排查 RISC-V 专用路径性能回归时使用。

2026-07-31
support-new-llm
计算机系统分析师计算机程序员

为 MNN 框架添加新的 LLM 模型支持。支持从 HuggingFace/ModelScope 下载模型,分析架构,添加映射,Hook 对齐测试,导出 MNN 模型。采用 TDD 模式,分 6 步执行,每步有独立测试标准。

2026-04-09
test-ci
数据科学家

Run MNN tests / benchmarks on host or real devices. Covers two parallel tracks — (1) the regression / CI suite (static checks, host-side tests, on-device Android arm64 matrix via ./test.sh + test_stages.json) and (2) one-command iOS real-device LLM benchmarking (prefill/decode tok/s, branch comparison). Use when the user asks to run the tests, run CI, smoke-test a build, verify a change on a device, benchmark on-device (Android or iPhone/iPad), or add / select / retune a test stage.

2026-08-07
vulkan-optimize
计算机系统分析师

MNN Vulkan 后端 op/kernel 优化与扩展。覆盖 GLSL .comp + makeshader 双轨、conv1x1 dispatcher 多路径、coopMat/subgroup 路径、weight prepare/pack 流程、Adreno 真机稳定性。

2026-06-05
openspec-archive-change
计算机系统分析师

Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.

2026-02-14
openspec-bulk-archive-change
技术写作员网页与数字界面设计师

Archive multiple completed changes at once. Use when archiving several parallel changes.

2026-02-14
openspec-continue-change
技术写作员网页与数字界面设计师

Continue working on an OpenSpec change by creating the next artifact. Use when the user wants to progress their change, create the next artifact, or continue their workflow.

2026-02-14
openspec-explore
软件开发工程师计算机程序员

Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.

2026-02-14
openspec-ff-change
软件开发工程师

Fast-forward through OpenSpec artifact creation. Use when the user wants to quickly create all artifacts needed for implementation without stepping through each one individually.

2026-02-14
openspec-new-change
数据科学家

Start a new OpenSpec change using the experimental artifact workflow. Use when the user wants to create a new feature, fix, or modification with a structured step-by-step approach.

2026-02-14
openspec-onboard
软件开发工程师

Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.

2026-02-14
openspec-sync-specs
技术写作员

Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.

2026-02-14
openspec-verify-change
软件开发工程师网页与数字界面设计师

Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.

2026-02-14
openspec-apply-change
软件质量保证分析师与测试员

Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.

2026-02-14
open-code-review-delegate
市场经理

Delegation mode for open-code-review (OCR). Instead of OCR calling an LLM endpoint, this skill instructs the host agent to perform the code review itself, using OCR only for deterministic engineering: file selection and rule resolution. Use when the host agent should drive the review with its own LLM capabilities.

2026-08-07
open-code-review
市场经理

Performs AI-powered code review on Git changes using the `ocr` CLI from alibaba/open-code-review. Use when the user asks to review code, review a pull request, review staged/unstaged changes, review a commit, or compare branches for code quality issues. Produces line-level review comments and can automatically apply fixes when requested. With appropriate review rules, can detect various types of issues including bugs, security vulnerabilities, performance problems, and code quality concerns.

2026-06-24
open-code-review-delegate
计算机系统分析师

Delegation mode for open-code-review (OCR). Instead of OCR calling an LLM endpoint, this skill instructs the host agent to perform the code review itself, using OCR only for deterministic engineering: file selection and rule resolution. Use when the host agent should drive the review with its own LLM capabilities.

2026-08-07
open-code-review
网络与计算机系统管理员

Performs AI-powered code review on Git changes using the `ocr` CLI from alibaba/open-code-review. Use when the user asks to review code, review a pull request, review staged/unstaged changes, review a commit, or compare branches for code quality issues. Produces line-level review comments and can automatically apply fixes when requested. With appropriate review rules, can detect various types of issues including bugs, security vulnerabilities, performance problems, and code quality concerns.

2026-06-24
RuleHub — Agent Skills 市场与 AI 编程灵感