What is tilelang-pass-design?

A design-document skill for compiler passes in TileLang-Ascend. It plans where a pass belongs, what code representation it changes, how it should be implemented, how it should be tested, and what risks it has.

Use it when adding a new pass, deciding its pipeline position, documenting its input and output data, planning its implementation, or defining its tests.

It resolves the important architectural decisions before code is written, reducing the chance of placing a pass incorrectly or missing its dependencies.

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-pass-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-pass-design/github.svg)](https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-pass-design)

Your own site

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

Per session 98 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 2,918 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	$0.00098	$0.02918
Opus 5.5	$0.00039	$0.01167
Sonnet 5.5	$0.00020	$0.00584
Haiku 4.5	$0.00010	$0.00292

Measured 2mo ago against content hash c3089b5a849a, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

tilelang-pass-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-pass-design/SKILL.md · 261 lines

How it starts

The opening of the file, as written. The whole thing — 261 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TileLang-Ascend Pass 设计文档生成 Skill

1. 目标

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

Pass 定位：Phase 1 / Phase 2 归属、Pipeline 具体位置
依赖分析：上游 Pass 输入、下游 Pass 输出、数据流传递
IR 变换设计：输入 IR 结构、输出 IR 结构、变换逻辑
实现方案：C++ 类设计、核心方法、Python Wrapper、Pipeline 集成
测试方案：功能测试、依赖测试、边界测试
风险分析：已知约束、常见错误、与其他 Pass 的交互

2. 输入要求

必需信息

字段	说明
Pass 名称	如 `BufferReuseOptimizer`、`L0CLayoutOptimization`
功能描述	Pass 要解决的问题和目标
IR 变换类型	修改 IR / 收集信息 / 验证 IR
平台范围	平台无关 / Ascend 特定

提问规则（必须严格遵守）：

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：Pass 名称 → 功能描述 → IR 变换类型 → 平台范围
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

字段	说明
参考 Pass	可参考的现有 Pass 名称
输入数据依赖	Pass 需要哪些 attrs（如 `buffer_shapess`、`address_map`）
输出数据供给	Pass 产生哪些 attrs 供下游使用
性能目标	Pass 对编译时间或生成代码性能的影响

3. 工作流程

Phase 1：需求澄清

解析用户提供的 Pass 需求信息
检查必需字段是否完整
按顺序逐一提问补全缺失字段（每次只问一个）

Phase 2：信息收集

查阅 Pass 定位参考资料：
- tilelang-pass-workflow-analyzer/references/pass-pipeline-overview.md - Pipeline 架构
- tilelang-pass-workflow-analyzer/references/new-pass-placement-guide.md - 定位决策流程
- tilelang-pass-workflow-analyzer/references/pass-dependency-graph.md - 依赖关系
查阅 Pass 实现参考资料：
- tilelang-pass-analyzer/references/pass-registry-ascend.md - 类似 Pass 实现
- tilelang-pass-analyzer/references/ir-examples.md - IR 变换示例格式
查阅本 skill 的实现模式参考：
- references/pass-impl-patterns.md - C++ 类模板、注册方式

Phase 3：Pass 定位分析

按照 new-pass-placement-guide.md 的决策流程：

Step 1：分析 Pass 功能

Pass 的主要功能是什么？
Pass 修改 IR 还是收集信息？
Pass 是平台无关还是 Ascend 特定？
Pass 属于哪类优化（Lowering / 合法化 / 内存 / 流水线 / 同步 / 其他）？

Step 2：分析依赖关系

Pass 需要哪些输入数据（attrs）？
这些数据由哪个 Pass 产生？
Pass 产生哪些输出数据？
这些数据由哪个 Pass 使用？

Step 3：确定阶段归属

DSL Lowering / IR 合法化 → Phase 1
硬件优化 / 内存优化 / 同步优化 → Phase 2
输入数据来自 Phase 1 → 可在 Phase 1 或 Phase 2
输入数据来自 Phase 2 → 必须在 Phase 2

Read the full file on GitHub · 261 lines

Files

What ships with it

2 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 · 261 lines · 98 tokens per session scan A c3089b5a849a

Subscribe to this mod's changes

tilelang-pass-design is a skill published in the GitHub repository tile-ai/tilelang-ascend (427 stars, last pushed yesterday), licensed MIT. It adds 98 tokens to every session and 2,918 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.

Planning

tilelang-pass-design next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
tilelang-pass-designtile-ai/tilelang-ascend	Skill	427	yesterday	MIT	A	98	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