TypingMindTypingMindThe best frontend for AI
Ascend Transformer Boost logo

Ascend Transformer Boost

Organization
Ascend
ascend-transformer-boost

昇腾 Transformer 加速库（ATB）核心技能集索引（Index Skill）。 整合 8 大核心技能：CANN 安装部署、ATB 测试框架编译、 ATB→ACLNN 算子替换设计文档生成、ATB→ACLNN 算子迁移，覆盖昇腾 NPU 开发全链路。

Overview

PublisherAscend
Repositoryagent-skills
Skill nameascend-transformer-boost
Stars
54
Forks
4
Bundled files
8
LicenseMulanPSL-2.0
Links
Skill sourceRepository
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 8 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Ascend on GitHub. Read the source before you install it.

Installation

Install the Ascend Transformer Boost AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Ascend/agent-skills/tree/master/skills/ascend-transformer-boost
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Ascend/agent-skills.git /tmp/agent-skills
mkdir -p .claude/skills
cp -r /tmp/agent-skills/skills/ascend-transformer-boost .claude/skills/ascend-transformer-boost
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ascend Transformer Boost in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Ascend Transformer Boost on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Ascend Transformer Boost is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

ascend-transformer-boost 技能集

Quick Reference

用途: ATB（Ascend Transformer Boost）开发全流程技能索引，从环境搭建到算子迁移。

核心命令:

bash
# 迁移标准化工作流
atb-ops-to-aclnn-migration-workflow

调用时机

当用户提到以下关键词时触发此索引：

  • ATB / ascend-transformer-boost / ascend transformer boost
  • 昇腾 + 开发环境 / NPU
  • ATB + ACLNN / 算子迁移
  • CANN + 安装 / 910B / 950
  • ATB + CSV / 测试 / 调试

详细内容

技能索引表

技能目录名称类型核心功能
atb-ops-to-aclnn-migration-workflow/OPS→ACLNN 迁移标准化工作流workflow-template7 阶段标准化工作流模板，含 HIL 用户确认机制
atb-nnal-installer/CANN NNAL 安装env-setupNNAL 安装（Toolkit+Kernels 由 cann-operator-env-config 提供）
atb-testframework-build/ATB 测试框架编译build全量编译 / 增量编译，GitHub→gitcode 镜像源替换
atb-aclnn-operator-replacement-designer/ATB→ACLNN 算子替换设计文档生成design输入 ATB/ACLNN 接口链接，生成 7 章结构化设计文档，含 HIL 用户确认
atb-aclnn-operator-migration/ATB→ACLNN 算子迁移工具migration执行算子迁移，910B/950 设备启用 ACLNN 加速
atb-csv-tester/ATB CSV 测试执行test运行 CSV 格式 ATB 测试用例
atb-csv-testcase-generator/ATB CSV 测试用例生成testcase正例/反例/性能测试设计，覆盖率分析，含 HIL 用户确认
atb-debug-guide/ATB 调试指南debug环境问题排查、ABI版本、内存错误、CSV测试失败

技能依赖关系

ascend-transformer-boost (Index Skill)
│
├── atb-ops-to-aclnn-migration-workflow         ← 标准化工作流模板
│       │
├── atb-nnal-installer                 ← 依赖 cann-operator-env-config（提供 Toolkit+Kernels）
│       │
├── atb-testframework-build             ← 依赖 atb-nnal-installer
│       │
├── atb-aclnn-operator-replacement-designer  ← 独立，迁移前必需
│       │ (含 HIL 用户确认 Gate 1)
│       │
├── atb-csv-testcase-generator          ← 依赖设计文档
│       │ (含 HIL 用户确认 Gate 2)
│       │
├── atb-aclnn-operator-migration        ← 依赖 atb-nnal-installer + 设计文档
│       │ (依赖 Gate 1 & Gate 2 通过)
│       │
├── atb-csv-tester                      ← 依赖 atb-nnal-installer + 测试框架
│       │
└── atb-debug-guide                    ← 辅助技能，独立使用

标准化工作流调用时序

[Phase 0] 前置学习
    │
    ▼
[Phase 1] 设计文档生成
    └─ atb-aclnn-operator-replacement-designer
        │
        └── [Gate 1: 用户确认] ⭐
                │
                ▼
[Phase 2] CSV用例设计
    └─ atb-csv-testcase-generator
        │
        └── [Gate 2: 用户确认] ⭐
                │
                ▼
[Phase 3] 实际迁移
    └─ atb-aclnn-operator-migration
        │
        ▼
[Phase 4] 编译验证
    └─ atb-testframework-build
        │ (失败) → atb-debug-guide
        ▼
[Phase 5a] 小范围验证（双文件）
    ├─ Step 1: atb-csv-tester（前台 _func.csv，正例+反例）
    └─ Step 2: subagent（后台 _perf.csv，逐个 -n N:N 避免 OOM）
        │ (失败) → atb-debug-guide
        ▼
[Phase 5b] 全量回归（汇合 subagent 结果）
    └─ atb-csv-tester (Gate 5 全量通过)
        │ (失败) → atb-debug-guide
        ▼
[Phase 6] 交付报告

依赖矩阵

调用技能前置技能前置条件
atb-ops-to-aclnn-migration-workflow无作为整体流程参考
atb-nnal-installercann-operator-env-configNPU 驱动已安装，Toolkit+Kernels 已由 cann-operator-env-config 安装
atb-testframework-buildatb-nnal-installerCANN 环境、ATB 源码、Docker 容器
atb-aclnn-operator-replacement-designer无ATB/ACLNN 接口文档链接
atb-csv-testcase-generator无ATB 源码、接口规格文档、设计文档已确认
atb-aclnn-operator-migrationatb-nnal-installer + replacement-designerCANN 环境、ATB 源码、设计文档、CSV 用例已确认
atb-csv-testeratb-nnal-installer + atb-testframework-buildCANN 环境、ATB 测试框架
atb-debug-guide无CANN 环境、ATB 测试框架

编译验证约束

当修改 ATB 算子源码（*_aclnn_runner.cpp、*_operation.cpp 等）后需要重新编译验证。

编译命令:

bash
# 在 Docker 容器内执行
source $ASCEND_TOOLKIT_HOME/set_env.sh
cd $ATB_REPO_PATH
bash scripts/build.sh testframework 2>&1 | tail -50

注意：确保环境变量已设置：

  • ASCEND_TOOLKIT_HOME（CANN 安装路径）
  • ATB_REPO_PATH（ATB 仓库路径）

约束规则:

  • 最多验证 3 次：编译失败时记录原因，规划修改方案
  • Step by Step：每次只修复一个问题
  • 无法解决时：记录问题并询问用户

编译失败处理:

失败次数操作
第1次失败分析错误，定位问题，修复后重新编译
第2次失败深入分析，查看参考实现
第3次失败停止，记录错误和修改计划，询问用户

调用场景索引

场景推荐调用技能备注
安装 NNAL（ATB 加速库）atb-nnal-installer依赖 cann-operator-env-config 提供 Toolkit+Kernels
编译 ATB 测试框架atb-testframework-build需要 Docker + CANN 环境
为算子替换撰写设计文档atb-aclnn-operator-replacement-designer需要 ATB/ACLNN 接口链接
执行算子迁移代码atb-aclnn-operator-migration支持 910B/950 设备，建议先完成设计文档
运行 CSV 测试验证算子atb-csv-tester需要 CANN + 测试框架
编写算子 CSV 测试用例atb-csv-testcase-generator需要接口规格，覆盖 910B/950
ATB 调试/问题排查atb-debug-guide报错分析、环境配置

参考文档

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Ascend Transformer Boost AI skill do?

昇腾 Transformer 加速库（ATB）核心技能集索引（Index Skill）。 整合 8 大核心技能：CANN 安装部署、ATB 测试框架编译、 ATB→ACLNN 算子替换设计文档生成、ATB→ACLNN 算子迁移，覆盖昇腾 NPU 开发全链路。

Why use Ascend Transformer Boost on TypingMind?

Because you install it once and use it with any model. Ascend Transformer Boost is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Ascend Transformer Boost in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Ascend/agent-skills/tree/master/skills/ascend-transformer-boost. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Ascend Transformer Boost?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Ascend Transformer Boost?

As many as you like. As long as a model supports skills, you can use Ascend Transformer Boost with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Ascend Transformer Boost AI skill free?

Yes. It is published on GitHub by Ascend under the MulanPSL-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

Related AI Skills

View all
openclaw logo

Openclaw Refactor Docs

openclaw
OrganizationPopular

Refactor an existing OpenClaw docs page with source-audited preservation, restructuring, and verification.

82.2K
391K
Instructions
View
openclaw logo

Openclaw Repair Sweep

openclaw
OrganizationPopular

Run scoped OpenClaw issue/PR repair campaigns: coordinate workers, prove root causes, and land or close verified work under the requested authority.

82.2K
391K
1 files
View
openclaw logo

Openclaw Test Performance

openclaw
OrganizationPopular

Benchmark, diagnose, and optimize OpenClaw test and plugin-suite runtime, import hotspots, CPU/RSS, heap growth, and slow coverage paths.

82.2K
391K
1 files
View
openclaw logo

Openclaw Testing

openclaw
OrganizationPopular

Choose proportional OpenClaw tests and checks, diagnose failures, and route environment-sensitive or release proof to its owner.

82.2K
391K
2 files
View

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇