What is tilelang-a5-sim-convert?

A converter that creates a separate script for running a TileLang example in the A5 camodel simulator, a software environment that imitates Ascend hardware execution.

Use it when converting an example for A5 simulation, preparing a camodel run, or testing a kernel in simulation mode instead of on an NPU.

It lets developers test a kernel without running it on a physical NPU. The original script remains unchanged while the simulator version is generated alongside it.

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-a5-sim-convert/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-a5-sim-convert/github.svg)](https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-a5-sim-convert)

Your own site

<a href="https://agentmods.dev/skills/tile-ai/tilelang-ascend/tilelang-a5-sim-convert"><img src="https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-a5-sim-convert/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-a5-sim-convert"><img src="https://agentmods.dev/badge/skills/tile-ai/tilelang-ascend/tilelang-a5-sim-convert.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 93 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,864 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	$0.00093	$0.01864
Opus 5.5	$0.00037	$0.00746
Sonnet 5.5	$0.00019	$0.00373
Haiku 4.5	$0.00009	$0.00186

Measured 2mo ago against content hash 8025c7c38594, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

tilelang-a5-sim-convert 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.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/parse_example.py, scripts/run_a5_sim_template.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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-a5-sim-convert/SKILL.md · 173 lines

How it starts

The opening of the file, as written. The whole thing — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TileLang A5 Camodel 仿真脚本转换

将任意 tilelang DSL 脚本转换为 A5 camodel 仿真可运行的独立脚本。

模板结构（260 行，只改两处）

模板文件：.agents/skills/tilelang-a5-sim-convert/scripts/run_a5_sim_template.py

行 1-24    import 语句            ← 不动
行 25-96   环境自动设置            ← 不动（_find_ascend_home, _source_cann, _find_sim_lib, setup）
行 99-133  加载 camodel 运行时    ← 不动（load_runtime, dev_malloc）
行 136-166 kernel 定义            ← ★ 第 1 处要改
行 169-260 main() 编译+运行+验证   ← 部分要改（详见下方）

工作流程

收到脚本路径后，按以下步骤执行：

Step 1: 运行解析脚本获取 kernel 信息

cd <tilelang-ascend-root>
python .agents/skills/tilelang-a5-sim-convert/scripts/parse_example.py <target_script>

输出 JSON，包含 kernel_name、buffers（shape/dtype 列表）。

Step 2: 读取模板 + 原始脚本

读取 .agents/skills/tilelang-a5-sim-convert/scripts/run_a5_sim_template.py
Read 目标脚本，找到 kernel 定义部分（@T.prim_func 或 @tilelang.jit 装饰的函数体）

Step 3: 生成 *_sim.py

输出路径：<原路径>/<原名>_sim.py（绝不覆盖原始文件）。

改动清单

改动 1：kernel 定义（模板 136-166 行）

原始脚本	仿真脚本
`@tilelang.jit(out_idx=[-1])`	删掉
`def matmul(M, N, K, ...):`	`def make_kernel():`
`T.Tensor((M, K), dtype)`	`T.Tensor((1024, 256), "float16")` ← 用 Step1 解析出的具体数值
`T.alloc_L0C(..., "float16")`	`T.alloc_L0C(..., "float")` ← A5 pto-isa 要求 float32
`func = matmul(...)` 触发编译	删掉，编译在 main() 里统一处理

生成的代码结构：

def make_kernel():
    import tilelang.language as T
    @T.prim_func
    def main(
        A: T.Tensor((1024, 256), "float16"),   # ← 具体数值
        B: T.Tensor((256, 1024), "float16"),
        C: T.Tensor((1024, 1024), "float16"),
    ):
        # ... kernel 逻辑（和原始脚本一模一样）...
    return main

改动 2：数据准备（模板 214-233 行）

a) 维度变量（第 215 行）

根据 Step1 的 buffers 设置：

# 原始模板（gemm 专用）
M, N, K = 1024, 512, 256

# 通用写法：从 buffers 提取
# buffers[0].shape = [M, K]  →  M = shape[0], K = shape[1]
# buffers[1].shape = [K, N]  →  N = shape[1]
# buffers[2].shape = [M, N]

如果不是矩阵（比如 1D/3D tensor），按实际 shape 处理。

b) 数据 dtype（第 216-218 行）

Read the full file on GitHub · 173 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 · 173 lines · 93 tokens per session scan A 8025c7c38594

Subscribe to this mod's changes

tilelang-a5-sim-convert is a skill published in the GitHub repository tile-ai/tilelang-ascend (427 stars, last pushed yesterday), licensed MIT. It adds 93 tokens to every session and 1,864 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.

Hardware and IoT

tilelang-a5-sim-convert next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
tilelang-a5-sim-converttile-ai/tilelang-ascend	Skill	427	yesterday	MIT	A	93	Claude Code, Codex
gke-compute-classesgoogle/skills	Skill	22k	yesterday	Apache-2.0	A	83	Claude Code, Codex
jetson-diagnosticNVIDIA/skills	Skill	3.9k	yesterday	Apache-2.0	A	30	Claude Code, Codex
holoscan-install-wheelNVIDIA/skills	Skill	3.9k	yesterday	Apache-2.0	A	34	Claude Code, Codex
offensive-z-waveSnailSploit/Claude-Red	Skill	10k	25d ago	MIT	A	113	Claude Code, Codex
doca-socket-relayNVIDIA/skills	Skill	3.9k	yesterday	Apache-2.0	A	236	Claude Code, Codex
hsb-flashNVIDIA/skills	Skill	3.9k	yesterday	Apache-2.0	B	94	Claude Code

tilelang-ascend: Skill for Claude Code