Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/slowlyC/agent-gpu-skillsnpx agentmods add skills/slowlyc/agent-gpu-skills/cutlass-skillWrote 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/slowlyc/agent-gpu-skills/cutlass-skill)<a href="https://agentmods.dev/skills/slowlyc/agent-gpu-skills/cutlass-skill"><img src="https://agentmods.dev/badge/skills/slowlyc/agent-gpu-skills/cutlass-skill.svg" alt="Measured on agentmods" 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 | $0.00137 | $0.01356 |
| Opus 5 | $0.00068 | $0.00678 |
| Sonnet 5 | $0.00027 | $0.00271 |
| Haiku 4.5 | $0.00014 | $0.00136 |
Grade A, and why
cutlass-skill 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 8d 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 — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CUTLASS and CuTe development
Use the local CUTLASS checkout as the primary implementation reference. Prefer current source and examples over remembered APIs because CuTeDSL paths and interfaces change frequently.
Locate the checkout
Resolve the directory containing this SKILL.md, then use its repos/cutlass/ child. The installer links that path to agent-gpu-skills/third_party/cutlass/ or to the checkout supplied through CUTLASS_REPO.
In commands below, replace CUTLASS_REPO with the resolved absolute path:
CUTLASS_REPO=/absolute/path/to/cutlass-skill/repos/cutlass
If the checkout is missing, run this from the agent-gpu-skills repository and reinstall the Skill:
bash update-repos.sh cutlass
bash install.sh --skill cutlass-skill
Choose the source surface
| Task | Start here |
|---|---|
| CuTeDSL API and implementation | python/CuTeDSL/cutlass/ |
| All CuTeDSL examples and tutorials | examples/python/CuTeDSL/ |
| CUTLASS C++ examples | examples/ |
| CUTLASS C++ kernels and builders | include/cutlass/ |
| CuTe C++ layout, copy, and MMA | include/cute/ |
| Python layout and swizzle utilities | python/pycute/ |
| Python manifest and generator utilities | python/cutlass_library/ |
Read quick-reference.md when mapping an operation or architecture to a concrete file in the validated checkout.
Query workflow
- Identify the programming surface: CuTeDSL Python, CuTe C++, or CUTLASS C++.
- Find the closest example for the target architecture and operation.
- Trace the example into the implementation and headers it instantiates.
- Check dtype, layout, alignment, architecture target, schedule, and epilogue together.
- Build or run the smallest representative example before adapting it to the target repository.
Discover files before loading a large source file:
find "$CUTLASS_REPO/examples/python/CuTeDSL" -type f | sort
find "$CUTLASS_REPO/examples" -mindepth 1 -maxdepth 1 -type d | sort
What ships with it
1 file 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.
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.
- 8d ago First seen · 144 lines · 137 tokens per session scan A 76a6af7df6a0
cutlass-skill is a skill published in the GitHub repository slowlyC/agent-gpu-skills (165 stars, last pushed 1mo ago), licensed MIT. It adds 137 tokens to every session and 1,356 once invoked, about $0.0007 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
cpp-pro
Writes, optimizes, and debugs C++ applications using modern C++20/23 features, template metaprogramming, and high-performance systems techniques. Use when building or refactoring C++ code requiring concepts, ranges, coroutines, SIMD optimization, or careful memory management — or when addressing performance…
memory-safety-patterns
Implement memory-safe programming with RAII, ownership, smart pointers, and resource management across Rust, C++, and C. Use when writing safe systems code, managing resources, or preventing memory bugs.
cpp
Comprehensive C/C++ programming reference covering everything from C11-C23 and C++11-C++23, system programming, CUDA GPU computing, debugging tools, Rust interop, and advanced topics. Use for: C/C++ questions, C/C++ interview preparation, modern language features, RAII/memory management, templates/generics, CUDA…
cpp-coroutines
Use when working with C++20 coawait, coyield, and coreturn, implementing promisetype, sizing coroutine frames, or debugging suspended coroutines in GDB. Not for Rust async: use idiomatic-rust.
assembly-riscv
Use when reading or writing RV32/RV64 assembly, inline asm in C, the RISC-V psABI, IMAFD extension naming, compressed instructions, or QEMU RISC-V debugging.
compiler-optimizations-deep
Use when -O3 leaves a hot loop scalar, spills appear in assembly, or a PGO or BOLT deployment is planned or stalls. Not for machine lowering: use code-generation-and-backends.