What is tilelang-op-design?

A design-document generator for TileLang Ascend operators, where an operator is a computation such as softmax, layer normalization, or matrix multiplication.

Use it when designing a new operator, planning a migration, or generating a design.md file with the programming mode, API mapping, memory layout, tiling strategy, and verification plan.

It turns an operator’s formula and tensor shapes into a documented implementation plan. This makes choices about memory, tiling, loops, synchronization, and testing explicit before coding.

How tile-ai/tilelang-ascend configures Claude Code and Codex

Reuse

Borrowing it

Nothing to install: this file belongs to tile-ai/tilelang-ascend. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/tile-ai/tilelang-ascend/ascendc_pto/.agents/skills/tilelang-op-design/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/tile-ai/tilelang-ascend

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-op-design/github.svg)](https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-op-design)

Your own site

<a href="https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-op-design"><img src="https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-op-design/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-op-design"><img src="https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-op-design.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 92 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,188 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What 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 5.1	$0.00092	$0.03188
Opus 5.5	$0.00037	$0.01275
Sonnet 5.5	$0.00018	$0.00638
Haiku 4.5	$0.00009	$0.00319

Measured 2mo ago against content hash 1df54d4c8494, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

tilelang-op-design 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 2mo 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.

.agents/skills/tilelang-op-design/SKILL.md · 191 lines

How it starts

The opening of the file, as written. The whole thing — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TileLang-Ascend 算子设计文档生成

1. 目标

根据算子需求信息，生成一份完整的 TileLang-Ascend 算子设计文档（design.md），涵盖以下核心决策：

编程模式选型：Developer / Expert / 混合模式
API 映射：将数学公式拆解为 TileLang DSL 原语组合
内存层级规划：GM → L1/UB → L0 的数据搬运路径
Tiling 策略：Block 划分与 Tile Shape 设计
循环结构：T.Parallel / T.serial / T.Pipelined / T.Persistent 的选择
同步策略：自动同步 vs 手动同步标志
验证方案：Golden 函数与 L0 门槛测试计划（完整分层套件 L1/L2/Boundary 由 tilelang-op-test-design 生成）

2. 输入要求

必需信息

字段	说明
算子名称	如 `softmax`、`layer_norm`、`flash_attention`
数学公式	算子的数学表达，如 $\text{softmax}(x_i) = e^{x_i} / \sum e^{x_j}$
输入张量规格	shape、dtype
输出张量规格	shape、dtype
编程模式偏好	Developer / Expert / 混合
迁移算子路径 ⭐	原算子文件路径（迁移时必需），用于获取 golden 实现
输出形状 ⭐	原算子输出 shape（迁移时必需），如 `(N, M)` 或 `(M, N)`

迁移算子时必须提供原算子路径和输出形状，否则无法证明迁移正确性。Golden 实现一致性要求详见 tilelang-op-develop checklist.md #9 Golden 实现一致 / #10 输出形状匹配。

提问规则（必须严格遵守）：

优先使用调用方传入的字段：若调用方（如 @tilelang-op-orchestrator 通过 analyst 传入 op_requirements 结构）已经提供了字段值，全部跳过提问，直接进入技术约束检测和 design 生成
每次只询问一个字段：使用 question 工具时，questions 数组中只包含一个元素
按表格顺序依次询问：算子名称 → 数学公式 → 输入张量规格 → 输出张量规格 → 编程模式偏好
已提供的字段跳过：如果用户在初始请求中已提供某个字段的值，跳过该字段继续下一个
示例：
- 第 1 次询问：只问"数学公式"
- 用户回答后，第 2 次询问：只问"输入张量规格"
- 以此类推

⚠️ 当被 orchestrator → analyst Subagent 链路调度时：

analyst 会把 orchestrator 在 Primary 上下文预检收集到的 op_requirements 完整传入
此时 5 个必需字段应当全部已 provided，跳过整个提问环节
若 skill 仍发现字段歧义或缺漏，不要在当前 Subagent 上下文调用 AskUserQuestion（透传不到真实用户），而是让 analyst 返回 partial_input + 缺失字段名给 orchestrator，由 orchestrator 在 Primary 上下文追问

字段	说明
典型配置	常用的 shape 组合与优先级
参考实现	PyTorch / NumPy 参考代码
性能目标	目标吞吐量或延迟
动态轴说明	哪些维度在运行时变化

3. 技术约束（必须遵守）

本项目为 TileLang-Ascend（华为昇腾 NPU），与 GPU 版 TileLang 有显著差异。外部参考实现不可直接使用，必须转换为 Ascend 兼容方案。

生成 design.md 前必须执行强制检测：三维 Kernel、threads 参数、动态循环边界、GPU 专用 API、GEMM 非整除、L0C 溢出等。

Read the full file on GitHub · 191 lines

Files

What ships with it

6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

2mo ago First seen · 191 lines · 92 tokens per session scan A 1df54d4c8494

Subscribe to this mod's changes

tilelang-op-design is a skill published in the GitHub repository tile-ai/tilelang-ascend (427 stars, last pushed yesterday), licensed MIT. It adds 92 tokens to every session and 3,188 once invoked, about $0.0004 per session on Opus 5.5. 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.

tilelang-op-design next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
tilelang-op-designtile-ai/tilelang-ascend	Skill	427	yesterday	MIT	A	92	Claude Code, Codex
imagegenopenai/codex	Skill	133k	yesterday	Apache-2.0	A	113	Codex
playwright-climicrosoft/playwright	Skill	99k	yesterday	Apache-2.0	A	19	Claude Code
local-ai-agentsmicrosoft/ai-agents-for-beginners	Skill	79k	4d ago	MIT	A	200	Claude Code, Codex
systematic-debuggingobra/superpowers	Skill	313k	3d ago	MIT	A	21	Claude Code
next-cache-components-adoptionvercel/next.js	Skill	143k	today	MIT	A	95	Claude Code
next-partial-prefetching-adoptionvercel/next.js	Skill	143k	today	MIT	A	103	Claude Code

tilelang-ascend: Skill for Claude Code