tilelang-ascend: Skill for Claude Code

.agents/skills/tilelang-ascend-tile-api/SKILL.md

tilelang-ascend-tile-api is a skill for Claude Code, Codex from tile-ai/tilelang-ascend. It costs 96 tokens per session (2,619 once invoked), scanned A, original, MIT.

A development process for adding new TileLang APIs for Huawei Ascend AI processors. It covers the user-facing Python API and the lower-level code that turns it into Ascend C code.

In plain words
What is it for?
It is for creating, exposing, implementing, or testing new T.tile.xxx operations in TileLang, including their compiler support and CI tests.
Why use it?
It helps prevent an API from being added only at the surface level without working compilation, generated code, documentation, and tests.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents); mentions AGENTS.md.

This is tile-ai/tilelang-ascend's own configuration. It tells Claude Code and Codex how to work on tilelang-ascend itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything tilelang-ascend configures →

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-ascend-tile-api/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.

agentmods badge for tilelang-ascend-tile-api

README.md
[![agentmods](https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-ascend-tile-api/github.svg)](https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-ascend-tile-api)
Your own site
<a href="https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-ascend-tile-api"><img src="https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-ascend-tile-api/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.

agentmods 80×15 button for tilelang-ascend-tile-api

Your own site · 80×15
<a href="https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-ascend-tile-api"><img src="https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-ascend-tile-api.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 96 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,619 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 warn 7 Sept 2026
SkillSpector: 3 findings, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Agent Snooping · line 17
    Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
    Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
  • medium Agent Snooping · line 19
    Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
    Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
  • medium Agent Snooping · line 139
    Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
    Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00096 $0.02619
Opus 5.5 $0.00038 $0.01048
Sonnet 5.5 $0.00019 $0.00524
Haiku 4.5 $0.00010 $0.00262

Measured 1mo ago against content hash 74641ab4a1e0, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

tilelang-ascend-tile-api 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 1mo 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-ascend-tile-api/SKILL.md · 199 lines

How it starts

The opening of the file, as written. The whole thing — 199 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TileLang Ascend Tile API 开发流程

当任务是新增一个面向用户的 T.tile.xxx 小 API 时使用本 skill,尤其是从 tilelang/language/ascend_tile.py 暴露、并由 Ascend C 代码生成支撑的 API。

这里的目标不只是增加一个 Python 函数名,而是交付一个真正可用的 API:它能编译、能 lowering、能生成合法的 Ascend C、有清晰稳定的语义边界,并且有合适的 CI 测试覆盖。

第一轮调研

开始实现前,先读清楚当前仓库形态,再决定方案:

  1. 阅读 AGENTS.md。
  2. 阅读 .agents/skills/tilelang-custom-skill/tilelang-api-best-practices/SKILL.md。
  3. 阅读 .agents/skills/tilelang-custom-skill/tilelang-api-best-practices/references/api-compute.md。
  4. 如果涉及编程模式或 pass_configs,阅读 .agents/skills/tilelang-custom-skill/tilelang-programming-model-guide/SKILL.md。
  5. 查看 tilelang/language/ascend_tile.py 中最相近的现有 API。
  6. 查看 testing/python/language/ 中最相近的测试。
  7. 查看 src/op/ascend.{h,cc}、src/target/codegen_ascend.cc、src/tl_templates/ascend/common.h 中相近的 lowering、codegen 和 helper 实现。

不要凭记忆推断 API 签名。已有本地模式优先于看起来更聪明的新抽象。

范围决策

编辑前,先明确并说明 API 边界:

  • 用户 API 名称,例如 T.tile.foo(dst, src, ...)。
  • 它属于纯 tile 计算、数据搬运、类 reduction 行为,还是带副作用的写回。
  • 支持的 buffer scope,通常是 GM、UB(shared.ub)、L1(shared.l1)、L0(wmma.*),或其中子集。
  • 支持的 dtype 和 rank。
  • 是否接受 Buffer、BufferLoad、BufferRegion。
  • 不支持的参数组合和语义。
  • 应该支持 Developer pass_configs、Expert 模式,还是两者都支持。

当语义无法和主仓 / GPU 的全局 API 对齐时,优先新增 Ascend 专属的 T.tile.xxx API。除非用户明确要求、且语义确实一致,否则不要新增或修改全局 T.xxx API。

实现路径

1. Python 前端

在 tilelang/language/ascend_tile.py 中新增用户入口。

优先复用该文件和 tilelang/language/copy_op.py 里的本地 helper 模式:

  • 如果现有 API 会解析 let-bound value,新 API 也应保持一致。
  • 只在语义明确时接受 Buffer、BufferLoad 或 BufferRegion。
  • 当 C++ lowering 需要 region、rank 或 extent 信息时,把前端输入转换为 tl.region。
  • 对不支持的 scope 或参数组合尽早报错,并给出清晰错误信息。
  • 当操作需要 C++ lowering 时,发射命名清晰的 op,通常是 tl.ascend_<api_name>。

除非操作本质上就是普通 copy,否则不要为了省事复用 tl.ascend_copy。有副作用、会改变硬件模式、或语义不同的操作,通常应该有显式 op 名称。

2. C++ Operator 和 Lowering

如果前端发射新的 tl.ascend_* op,需要补齐对应 C++ operator 路径:

  • 在 src/op/ascend.h 中声明。
  • 在 src/op/ascend.cc 中实现参数解析和 lowering。
  • 当它需要 tile-op lowering 时,用 TIR_REGISTER_TL_OP 注册。
  • 保留 region 信息,直到 lowering 可以计算访问指针、extent、stride、mask 或合法 shape。
  • 即使 Python 已经做过校验,C++ 侧也要重新检查关键约束。

Read the full file on GitHub · 199 lines

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.

  1. 1mo ago Changed · -37 lines 74641ab4a1e0
  2. 1mo ago Changed · +4 lines 30564c025b5d
  3. 1mo ago Changed · +39 lines cba02f138a1e
  4. 2mo ago First seen · 193 lines · 96 tokens per session scan A 5330c4b4ce70

Subscribe to this mod's changes

tilelang-ascend-tile-api is a skill published in the GitHub repository tile-ai/tilelang-ascend (427 stars, last pushed yesterday), licensed MIT. It adds 96 tokens to every session and 2,619 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.

Related

tilelang-ascend-tile-api compared with skills from other repositories

tilelang-ascend-tile-api next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
tilelang-ascend-tile-apitile-ai/tilelang-ascendSkill427yesterdayMITA96Claude Code, Codex
doca-argpNVIDIA/skillsSkill3.9kyesterdayApache-2.0A277Claude Code, Codex
holoscan-module-lifecycleNVIDIA/skillsSkill3.9kyesterdayApache-2.0A47Claude Code, Codex
embedded-stm32Mindrally/skillsSkill29729d agoApache-2.0A87Claude Code
embedded-firmware-devAmethystLuna/embedded-workbenchSkill30yesterdayMITA59Claude Code
hls-c2rtlvibeic/vibe-icSkill32todayApache-2.0A79Claude Code
hip-kernel-optimizationAMD-AGI/ApexSkill781mo agoMITA56Claude Code, Codex