Important
Tech Preview: Catalog under active development. We’re building the catalog in the open, sharing progress as the foundations take shape. Expect frequent changes as skills, categories, and descriptions evolve.
AMD Skills provide agents with knowledge, scripts, and conventions for working with AMD hardware and software.
Skills in this repository follow the standardized Agent Skills format and are designed to interoperate with the major coding agents like Cursor, Claude Code, OpenAI Codex, and Gemini CLI.
AMD Skills will soon be installable directly in Claude/Cursor/Codex and other agents through marketplace integration.
While we work marketplace integration, please refer to our Manual installation steps.
A skill is a self-contained folder that bundles everything an agent needs to perform a focused task: instructions, helper scripts, prompts, templates, and references. At its core is a SKILL.md file with YAML frontmatter, a name, and a short description that tells the agent when the skill should activate, followed by the guidance the agent reads while the skill is in use.
skills/
rocm-doctor/
SKILL.md
skill-card.md
scripts/
references/
When an agent decides a skill is relevant (or you invoke it explicitly), it loads that SKILL.md and follows the instructions inside. Descriptions stay in context cheaply; the full body of a skill only loads when the task actually matches.
Every skill also ships a skill-card.md: a short, human-facing governance card (Description, Owner, License) that tells a reviewer what the skill is and who stands behind it without reading the source. See docs/skill-cards.md.
Documentation describes an API surface: every flag, every option, neutral by design. A skill encodes the opinionated path: which flags, which container image, which gfx target, which environment variables, in what order. It captures the decisions a senior AMD engineer makes without thinking, in a form the agent can apply consistently across teams and repositories.
Skills earn their keep on repeated, opinionated workflows, exactly where the AMD stack lives.
The initial catalog is organized into three focus areas, spanning the full stack from client to cloud. This catalog is expected to grow significantly as more skills land.
Run and optimize on Ryzen AI.
| Skill | What it does | Source |
|---|---|---|
local-ai-use |
Route image generation, text-to-speech, and speech-to-text through a local AI server to reduce token cost. | in-repo |
local-ai-app-integration |
Integrate local AI into cloud LLM apps for offline support, better privacy, and lower API costs. | in-repo |
apu-memory-tuner |
Inspect and tune the shared-vs-dedicated memory split (GTT / UMA Frame Buffer) on AMD Ryzen APUs. | planned |
Cross-stack skills, from client to cloud.
| Skill | What it does | Source |
|---|---|---|
rocm-doctor |
Diagnose ROCm / PyTorch / llama.cpp failures on AMD GPUs against a fixed list of known misconfigurations. | planned |
llm-kernel-optimizer (hyperloom) |
Autonomously optimizes LLM inference on AMD GPUs. | planned |
vllm-semantic-router |
Setup a vLLM router that semantically maps your request to the best available platform. | planned |
Run and optimize on AMD Instinct.
| Skill | What it does | Source |
|---|---|---|
serving-llms-on-instinct |
Deploy LLM inference on AMD Instinct GPUs end-to-end: detect hardware (or onboard via AMD Developer Cloud), validate model fit, apply the right vLLM recipe, and launch a benchmarked endpoint. SGLang and engine/backend selection in later phases. | in-repo |
llm-kernel-evaluator (magpie) |
Evaluate GPU kernel correctness and performance, compare kernel implementations, and benchmark vLLM / SGLang inference with profiling, TraceLens, and torch-trace gap analysis. | Magpie |
The AMD stack is large and moves fast. ROCm, HIP, Ryzen AI, and framework integrations each have their own team, release cadence, and validation matrix. So skills here are federated: each skill is owned and versioned by the team that owns the product it describes, and this repository is the catalog that brings them together.
┌─────────────────────────────────────────────────────┐
│ amd/skills (this repo) │
│ │
│ skills/ .github/scripts/ .*-plugin/ │
│ in-repo skills sources.yml agent manifests │
└──────────────────────┬──────────────────────────────┘
│ one install
▼
your AI coding agent
▲
│ resolves pointers to
┌───────────────┬───────────────┼───────────────┬────────────────┐
│ │ │ │ │
ROCm/ROCm ROCm/HIP Ryzen AI repo lemonade-sdk ...more
rocm-doctor/ cuda-to-hip/ ryzen-ai-tools/ local-ai-app- product
gfx-target-... triton-amd-... ... integration/ repos
This repo also acts as an incubator: a skill can start under skills/ to iterate quickly, then graduate to its product repo and be re-pointed from .github/scripts/sources.yml once it has a clear owner, with no change for installed users.
skills/ # All skills the agent can load
docs/ # Long-form documentation (e.g. skill-cards.md)
.claude-plugin/ # Claude Code marketplace manifest
.cursor-plugin/ # Cursor marketplace manifest
plugin-metadata.json # Vendor-neutral identity/discovery metadata
.github/workflows/ # CI for validating skills
.github/scripts/ # Internal repo scripts
.github/scripts/sources.yml # External skill sources for federation
In-repo skills are authored directly under skills/. Federated skills are
declared in .github/scripts/sources.yml and vendored into
skills/ by the manually-dispatched import-external-skills workflow,
which opens a pull request with the imported copies. Each vendored skill
carries a .federated.json marker that records the upstream repo and
pinned commit, so the importer can refresh or remove it without disturbing
in-repo skills.
Until marketplace integration lands, install skills manually: clone this repo, then copy (or symlink) the skill folders you want from skills/ into your agent's skills directory. Each agent discovers SKILL.md automatically.
git clone https://github.com/amd/skills.git amd-skills
cp -r amd-skills/skills/local-ai-use <agent-skills-dir>/| Agent | Skills directory (personal / project) |
|---|---|
| Cursor | ~/.cursor/skills/ / .cursor/skills/ |
| Claude Code | ~/.claude/skills/ / .claude/skills/ |
| Codex | $HOME/.agents/skills / $REPO_ROOT/.agents/skills |
Once a skill is installed, reference it in plain language while talking to your agent. For example:
- "Use AMD Skills to integrate local AI capabilities into my app with Embeddable Lemonade."
- "Use AMD Skills to convert these CUDA kernels and flag anything that needs manual review."
In most cases the agent picks the right skill on its own from the description; explicit invocation is a fallback, not a requirement.
For hands-on, step-by-step guides that show a skill in action, see the walkthroughs.
We welcome contributions from AMD engineers and selected partners. Two paths, matching how the catalog is organized:
- Path A: In-repo skills. Authored directly under
skills/. Best for cross-cutting workflows without a natural product home. - Path B: Product-repo skills. Authored in a product repository and registered here through
.github/scripts/sources.ymlwith a pinned tag. Best for skills that should ship and version with a specific product.
See CONTRIBUTING.md for step-by-step instructions and the rules CI enforces.
Released under the MIT License. See LICENSE for details.
Copyright(C) 2026 Advanced Micro Devices, Inc. All rights reserved.