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.
curl -O https://raw.githubusercontent.com/tile-ai/tilelang-ascend/ascendc_pto/.agents/skills/tilelang-pass-analyzer/SKILL.mdgit clone --depth 1 https://github.com/tile-ai/tilelang-ascendWrote 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.
[](https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-pass-analyzer)<a href="https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-pass-analyzer"><img src="https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-pass-analyzer/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.
<a href="https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-pass-analyzer"><img src="https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-pass-analyzer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00154 | $0.03540 |
| Opus 5 | $0.00077 | $0.01770 |
| Sonnet 5 | $0.00031 | $0.00708 |
| Haiku 4.5 | $0.00015 | $0.00354 |
Grade A, and why
tilelang-pass-analyzer 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 383 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TileLang Pass 分析工具
⭐ 核心约束：分阶段信息获取
严格遵守以下信息获取优先级：
| 查询阶段 | 信息来源 | 工具使用 |
|---|---|---|
| 首次回答 | ✅ 优先使用 reference 文件 | read(references/*.md) |
| 用户追问深入细节 | ✅ 读取源码补充 | read(src/transform/*.cc) |
执行规则：
用户首次提问 Pass 功能
↓
1. 立即读取 reference 文件（pass-registry-*.md / pass-classification.md）
↓
2. 基于 reference 信息生成报告（足够回答 90% 的查询）
↓
3. 仅在用户追问"具体实现"、"源码细节"、"某个函数逻辑"时才读取源码
禁止行为：
- ❌ 首次回答就读取
src/transform/*.cc源码文件 - ❌ 忽略 reference 文件中的已有信息
- ❌ 过度深入分析辅助工具类（首次回答时）
- ❌ 首次回答就并行读取 reference 和源码
核心工作流程
Step 1: 解析用户意图
根据用户输入判断查询类型：
| 意图类型 | 关键词特征 | 处理方式 |
|---|---|---|
| 单 Pass 分析 | "XX pass 是做什么的"、"分析 XX pass"、"XX 的功能" | 查 registry → 分析代码 → 生成报告 |
| 双 Pass 对比 | "XX 和 YY 的区别"、"对比 XX YY"、"XX YY 的差异" | 生成对比报告（每节对比） |
| 双 Pass 联系 | "XX 和 YY 的联系"、"XX YY 的关系"、"XX YY 协作" | 生成联系报告（每节分析关系） |
| 分类查询 | "XX 类的 pass"、"内存相关 pass"、"Ascend pass 有哪些" | 查 classification → 列表报告 |
| 模糊匹配 | "sync 相关"、"vector pass"、"那个插入同步的" | 搜索 → 返回候选列表 |
Step 2: Pass 定位
精确匹配：
1. 读取 references/pass-registry-ascend.md 或 pass-registry-general.md
2. 根据名称获取：C++ 文件路径、Python 函数名、配置键
3. 定位到具体文件
模糊匹配策略：
1. 提取用户输入的关键词（去除 pass、transform 等通用词）
2. 在 registry 中搜索名称包含关键词的所有 pass
3. 返回候选列表（最多 5 个）：序号、名称、简述、分类
4. 提示用户选择序号或输入完整名称
分类查询：
1. 读取 references/pass-classification.md
2. 根据用户指定的分类（如"内存管理"、"同步"、"Ascend平台"）返回列表
3. 按子分类组织输出
Step 3: 代码分析（仅在用户追问深入时执行）
⚠️ 重要：首次提问时跳过此步骤，仅使用 reference 文件信息。
仅在用户追问以下问题时才执行此步骤：
- "某个函数的具体实现"
- "源码中的某个细节"
- "算法的具体逻辑"
- "某个配置项的作用"
按以下顺序分析 Pass 实现：
1. Python API 层
- 文件：
tilelang/transform/__init__.py - 提取：函数签名、docstring、参数
2. C++ 实现层
- 文件：
src/transform/<pass_name>.cc - 提取：
- 文件头注释
/*! \file ... */ - 主类名（通常继承
IRMutatorWithAnalyzer、StmtExprVisitor等） - 核心方法名（如
Substitute、VisitStmt、VisitExpr） - 注册宏
TVM_REGISTER_PASS_CONFIG_OPTION或TVM_REGISTER_GLOBAL
- 文件头注释
3. 配置层
- 文件：
tilelang/transform/pass_config.py - 提取：配置键名、默认值、描述
What ships with it
19 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.
- evals/evals.json 2.7 KB
- image.png 18 KB
- references/ir-examples.md 30 KB
- references/pass-classification.md 5.4 KB
- references/pass-designs/ascend_infer_buffer_scope_design.md 25 KB
- references/pass-designs/ascend_lower_parallel_to_vector_design.md 24 KB
- references/pass-designs/ascend_memory_planning_technical_doc.md 16 KB
- references/pass-designs/ascend_storage_rewrite_design.md 16 KB
- references/pass-designs/ascend_sync_insert_technical_doc.md 15 KB
- references/pass-designs/cross_core_pipeline_design.md 17 KB
- references/pass-designs/design_ascend_combinecv.md 19 KB
- references/pass-designs/image-1.png 43 KB
- references/pass-designs/image-2.png 116 KB
- references/pass-designs/image-3.png 51 KB
- references/pass-designs/image-4.png 67 KB
- references/pass-designs/pipeline_planning & inject_pipeline_design.md 17 KB
- references/pass-registry-ascend.md 7.6 KB
- references/pass-registry-general.md 5.6 KB
- T.pipelined.md 5.6 KB
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.
- 9d ago First seen · 383 lines · 154 tokens per session scan A c7f8a04e0b97
tilelang-pass-analyzer is a skill published in the GitHub repository tile-ai/tilelang-ascend (363 stars, last pushed yesterday), licensed MIT. It adds 154 tokens to every session and 3,540 once invoked, about $0.0008 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 skills, from other repositories
hr-onboarding
A new-hire onboarding plan as a single page — first week schedule, buddy + manager intro, learning track, equipment checklist, and "you're set when…" outcomes. Use when the brief mentions "onboarding", "new hire", "first week plan", or "入职".
book-mirror
Take any book (EPUB/PDF), produce a personalized chapter-by-chapter analysis. Each chapter is preserved in detail (The Chapter) and mirrored back to the reader's actual life (The Mirror) using brain context. The mirror observes and resonates — a friend pointing out parallels, NOT a consultant rearranging the reader's…
miniapp
Build a tiny interactive HTML playground only when someone asks to see, play with, or step through a mechanism.
eli5
Explain research, papers, or technical ideas in plain English with minimal jargon, concrete analogies, and clear takeaways. Use when the user says "ELI5 this", asks for a simple explanation of a paper or research result, wants jargon removed, or asks what something technically dense actually means.
deck-course-module
A course or workshop slide template with persistent learning goals, teaching pages, multiple-choice self-tests, and a wrap-up.
master-yinguang
A reference-based assistant for questions about Yinguang and Pure Land Buddhism, a Buddhist tradition focused on faith, ethical living, and practice connected with rebirth in the Pure Land. It can answer in Yinguang’s historical teaching style.