Agent skill

Llvm

Tune and analyse a FlyDSL kernel at the LLVM level: pick a compile hint, function attribute, or AMDGPU backend flag, then PROVE it reached codegen.

Custom licenceAuto-check: notesDevelopment

Install Llvm

skills CLI

$ npx skills add ROCm/FlyDSL --skill llvm -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install ROCm/FlyDSL llvm --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/ROCm/FlyDSL.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/llvm .claude/skills/llvm && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "llvm" agent skill from https://github.com/ROCm/FlyDSL/tree/main/.claude/skills/llvm into .claude/skills/llvm/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llvm", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/ROCm/FlyDSL/tree/main/.claude/skills/llvm

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add ROCm/FlyDSL --skill llvm -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install ROCm/FlyDSL llvm --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/ROCm/FlyDSL.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/llvm .agents/skills/llvm && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "llvm" agent skill from https://github.com/ROCm/FlyDSL/tree/main/.claude/skills/llvm into .agents/skills/llvm/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llvm", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add ROCm/FlyDSL --skill llvm -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install ROCm/FlyDSL llvm --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/ROCm/FlyDSL.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/llvm .cursor/skills/llvm && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "llvm" agent skill from https://github.com/ROCm/FlyDSL/tree/main/.claude/skills/llvm into .cursor/skills/llvm/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llvm", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/ROCm/FlyDSL.git --path .claude/skills/llvm

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add ROCm/FlyDSL --skill llvm -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install ROCm/FlyDSL llvm --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/ROCm/FlyDSL.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/llvm .gemini/skills/llvm && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "llvm" agent skill from https://github.com/ROCm/FlyDSL/tree/main/.claude/skills/llvm into .gemini/skills/llvm/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llvm", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install ROCm/FlyDSL llvm

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add ROCm/FlyDSL --skill llvm -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/ROCm/FlyDSL.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/llvm .github/skills/llvm && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "llvm" agent skill from https://github.com/ROCm/FlyDSL/tree/main/.claude/skills/llvm into .github/skills/llvm/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llvm", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add ROCm/FlyDSL --skill llvm -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install ROCm/FlyDSL llvm --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/ROCm/FlyDSL.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/llvm .opencode/skills/llvm && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "llvm" agent skill from https://github.com/ROCm/FlyDSL/tree/main/.claude/skills/llvm into .opencode/skills/llvm/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llvm", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: llvm
GitHub stars: 288
Token cost: ~4.9k tokens
SKILL.md length: 2,242 words
Files: 8 (incl. scripts, references)
Skills in repo: 19
Repo updated: Oct 8, 2026
First seen: Oct 7, 2026
Licence: Custom licence

At a glance

Tune and analyse a FlyDSL kernel at the LLVM level: pick a compile hint, function attribute, or AMDGPU backend flag, then PROVE it reached codegen.

Works in 10 steps: Establish your toolchain → Dump hygiene → The four lanes into LLVM → …
Asked to reduce spills
SKILL.md covers Pick the right skill first, Step 1 — Establish your…, Step 2 — Dump hygiene and Step 3 — The four lanes into…, plus 10 more sections
Runs Python scripts from its folder; calls docker and python

What it does

Llvm is an agent skill from ROCm/FlyDSL. Tune and analyse a FlyDSL kernel at the LLVM level: pick a compile hint, function attribute, or AMDGPU backend flag, then PROVE it reached codegen. Covers the compilehints keys, the llvmoptions cl::opt escape hatch, rocdl. / raw LLVM function attributes, per-arch options for gfx950 and gfx1250, dumping and reading NNllvmir.ll, replaying the LLVM half with llc, and the knobs in this repo that silently do nothing. Use when asked to reduce spills, raise occupancy, change the instruction schedule, set wavespereu or…

Its SKILL.md is about 4.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 9 other files, including scripts and reference files (for example `references/amdgpu-attributes.md`, `references/arch-gfx1250.md` and `references/arch-gfx950.md`).

It sits in Development, covering Performance optimization and Project scaffolding. The repository describes itself as: FlyDSL is the Python front‑end of the project: a Flexible Layout Python DSL for expressing tiling, partitioning, data movement, and kernel structure at a high level.

When your agent uses it

Asked to reduce spills
Raise occupancy
Change the instruction schedule
Flatworkgroupsize

Example prompts

“/llvm”

Requirements

Python 3
Docker
Pre-approved tools (allowed-tools): Read, Write, Edit, Bash, Grep, Glob

Workflow steps

10 steps, taken from the step headings in SKILL.md.

Establish your toolchain
Dump hygiene
The four lanes into LLVM
The compile_hints keys
Hint layering rules
Arch applicability
Verify it took effect
Knobs in this repo that do not work
Failure-mode taxonomy
Tuning workflow

What it can do on your machine

Read from SKILL.md and the folder at commit 1941889. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves these tools, so the agent can use them without asking each time:
- Read
- Write
- Edit
- Bash
- Grep
- Glob
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 2 files in scripts/ (Python), which the agent can run.
Shell commands in SKILL.md call:
- docker
- python
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use docker, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Llvm loads about 4.9k tokens when it runs, and up to ~14k if it reads all its reference files. Until then it costs about 235 tokens; SKILL.md has 2,242 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~235
When it runs · the whole SKILL.md, loaded when a task matches: ~4.9k
With references · SKILL.md plus every file in references/, read only if the agent opens them: ~14k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
```
allowed-tools: Read, Write, Edit, Bash, Grep, Glob
```

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 2,242 words (~4,918 tokens).

“FlyDSL reaches LLVM through four lanes, and only some of them are alive. This skill routes you to the one that works for your knob, then makes you prove it landed. The repo is knob-rich and instrument-poor: four knobs were…”

— opening of SKILL.md by ROCm, Custom licence

name: llvm
allowed-tools: Read, Write, Edit, Bash, Grep, Glob

Outline of SKILL.md

LLVM-Level Kernel Tuning and Analysis
Pick the right skill first
Step 1 — Establish your toolchain
Step 2 — Dump hygiene
Step 3 — The four lanes into LLVM
Step 4 — The compile_hints keys
Step 5 — Hint layering rules
Step 6 — Arch applicability
Step 7 — Verify it took effect
The mandatory negative control
Replaying just the LLVM half
Step 8 — Knobs in this repo that do not work
Step 9 — Failure-mode taxonomy
Step 10 — Tuning workflow
Analyzer scripts
References
Verified environment

Read the full SKILL.md on GitHub

Files

SKILL.md and 7 other files (scripts, references) in .claude/skills/llvm of ROCm/FlyDSL.

SKILL.md
references/amdgpu-attributes.md
references/arch-gfx1250.md
references/arch-gfx950.md
references/mllvm-flags.md
references/sched-primitives.md
scripts/llvm_ir_attrs.py
scripts/llvm_knob_check.py

Open the folder on GitHubat commit 1941889

Compare with similar skills

Llvm next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Llvm compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Llvm this skillROCm/FlyDSL	288	—	~4.9k	Automated safety check: Notes	Custom licence	today
Corvus Ctj Handler Implementationcorvus-dotnet/Corvus.JsonSchema	199	—	~4.4k	Automated safety check: Pass	Apache-2.0	today
Senior Javabenchflow-ai/skillsbench	1.8k	—	~5k	Automated safety check: Pass	MIT	2 mo ago
Code Review ChecklistshareAI-lab/learn-claude-code	78k	5 repos	~1.1k	Automated safety check: Pass	MIT	9 days ago
LLM Torch Profiler Analysissgl-project/sglang	37k	2 repos	~6.4k	Automated safety check: Pass	Apache-2.0	today
Nx Generatenomcopter/react-mosaic	4.8k	7 repos	~1.9k	Automated safety check: Pass	Custom licence	9 days ago

Similar skills

Corvus Ctj Handler Implementation
corvus-dotnet/Corvus.JsonSchema
Implement OpenAPI server handlers using Corvus.Text.Json generated types.
199 GitHub stars~4.4k tokensUpdated today
DevelopmentAuto-check passed
Senior Java
benchflow-ai/skillsbench
World-class Java and Spring Boot development skill for enterprise applications, microservices, and cloud-native systems.
1.8k GitHub stars~5k tokensUpdated 2 mo ago
Backend & APIsAuto-check passed
Code Review Checklist
shareAI-lab/learn-claude-code
Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.
78k GitHub starsUsed in 5 repos~1.1k tokens
DevelopmentAuto-check passed
LLM Torch Profiler Analysis
sgl-project/sglang
Unified LLM torch-profiler triage skill for sglang, vllm, TensorRT-LLM, and TokenSpeed.
37k GitHub starsUsed in 2 repos~6.4k tokens
DevelopmentAuto-check passed
Nx Generate
nomcopter/react-mosaic
Generate code using nx generators. An agent skill from nomcopter/react-mosaic.
4.8k GitHub starsUsed in 7 repos~1.9k tokens
DevelopmentAuto-check passed
Ponytail
DavidObando/gsharp
Forces the laziest solution that actually works, simplest, shortest, most minimal.
565 GitHub starsUsed in 8 repos~1.7k tokens
DevelopmentAuto-check passed

Questions about Llvm

What does Llvm do?

Tune and analyse a FlyDSL kernel at the LLVM level: pick a compile hint, function attribute, or AMDGPU backend flag, then PROVE it reached codegen. Llvm is an agent skill from ROCm/FlyDSL. Tune and analyse a FlyDSL kernel at the LLVM level: pick a compile hint, function attribute, or AMDGPU backend flag, then PROVE it reached codegen.

When should I use Llvm?

Llvm fits situations like: asked to reduce spills; raise occupancy; change the instruction schedule; flatworkgroupsize.

How do I install Llvm in Claude Code?

Run `npx skills add ROCm/FlyDSL --skill llvm -a claude-code`. Or copy the skill folder (.claude/skills/llvm in ROCm/FlyDSL) into .claude/skills/llvm in your project. Claude Code loads it when a task matches its description.

How do I install Llvm in Codex?

Run `npx skills add ROCm/FlyDSL --skill llvm -a codex`. Or copy the skill folder (.claude/skills/llvm in ROCm/FlyDSL) into .agents/skills/llvm in your project. Codex loads it when a task matches its description.

Can I use Llvm in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ROCm/FlyDSL --skill llvm -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/llvm, .gemini/skills/llvm, .github/skills/llvm and .opencode/skills/llvm in your project.

What does Llvm need to run?

Going by SKILL.md and its folder, Llvm needs Python for the scripts in its folder and the command-line tools its instructions call (docker and python). Our summary lists: Python 3; Docker. Its frontmatter pre-approves these tools: Read, Write, Edit, Bash, Grep, Glob.

Does Llvm access the network?

SKILL.md contains no URLs. Its commands use docker, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Llvm safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Llvm use?

Llvm has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.

How many tokens does Llvm use?

About 4.9k tokens (SKILL.md is roughly 20k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 9.4k tokens, read only when the agent opens those files.

What are the alternatives to Llvm?

Skills that share tags, products or a category with Llvm: Corvus Ctj Handler Implementation (corvus-dotnet/Corvus.JsonSchema, 199 stars), Senior Java (benchflow-ai/skillsbench, 1.8k stars), Code Review Checklist (shareAI-lab/learn-claude-code, 78k stars) and LLM Torch Profiler Analysis (sgl-project/sglang, 37k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Llvm?

ROCm (a GitHub organization) maintains it in ROCm/FlyDSL, which has 288 GitHub stars. The repository holds 19 skills in this directory. The repository was last updated on October 8, 2026.

Source: ROCm/FlyDSL on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Llvm

Install Llvm

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Files

Compare with similar skills

Similar skills

Corvus Ctj Handler Implementation

Senior Java

Code Review Checklist

LLM Torch Profiler Analysis

Nx Generate

Ponytail

More from ROCm/FlyDSL

Isa Resource Diff

API Stability

Build Rocm Image

Debug Flydsl Kernel

Flydsl Tile Programming

Format Code

Questions about Llvm

Install Llvm

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Files

Compare with similar skills

Similar skills

More from ROCm/FlyDSL

Related

Questions about Llvm