Tilegym Cutile Autotuning — Agent Skill by NVIDIA

From the author

Use when adding, modifying, optimizing, or debugging CuTile autotuning code. Trigger signals: `exhaustive_search` / `replace_hints` / `hints_fn` / `cuda.tile.tune` in code, `autotune` in filenames, or correctness/performance issues in autotuned CuTile kernels. Covers: tune-once/cache/launch pattern, per-architecture configs (sm80–sm120), parameter space design (tile sizes, occupancy, num_ctas), and 7 common pitfalls with solutions.

— the description by NVIDIA

Install

Get the skill from its source (we cannot redistribute it):
https://github.com/NVIDIA/skills/tree/fce3cca4170be75e82c64920654d56ee2811b005/skills/tilegym-cutile-autotuning

Documentation by NVIDIA

The licence of this skill does not allow us to copy its documentation. Read it at the source

Files

BENCHMARK.md(opens on GitHub)Not mirrored: opens on GitHub5.3 KiB
SKILL.md(opens on GitHub)Not mirrored: opens on GitHub18.3 KiB
assets/examples/01_rmsnorm_occupancy_only/autotuned_launch.py(opens on GitHub)ScriptNot mirrored: opens on GitHub7.9 KiB
assets/examples/01_rmsnorm_occupancy_only/fixed_launch.py(opens on GitHub)ScriptNot mirrored: opens on GitHub5.6 KiB
assets/examples/02_matmul_full_search/autotuned_launch.py(opens on GitHub)ScriptNot mirrored: opens on GitHub9.4 KiB
assets/examples/02_matmul_full_search/fixed_launch.py(opens on GitHub)ScriptNot mirrored: opens on GitHub5.4 KiB
assets/examples/03_rope_inplace_splitbuffer/autotuned_launch.py(opens on GitHub)ScriptNot mirrored: opens on GitHub11.3 KiB
assets/examples/03_rope_inplace_splitbuffer/fixed_launch.py(opens on GitHub)ScriptNot mirrored: opens on GitHub7.5 KiB
evals/evals.json(opens on GitHub)Not mirrored: opens on GitHub5.3 KiB
references/api-reference.md(opens on GitHub)Not mirrored: opens on GitHub7.8 KiB
references/hardware-constraints.md(opens on GitHub)Not mirrored: opens on GitHub11.1 KiB
references/kernel-type-templates.md(opens on GitHub)Not mirrored: opens on GitHub50.4 KiB
references/parameter-space-design.md(opens on GitHub)Not mirrored: opens on GitHub10.9 KiB
references/pitfalls.md(opens on GitHub)Not mirrored: opens on GitHub5.5 KiB
references/search-strategies.md(opens on GitHub)Not mirrored: opens on GitHub8.8 KiB
references/workflow.md(opens on GitHub)Not mirrored: opens on GitHub16.7 KiB
skill-card.md(opens on GitHub)Not mirrored: opens on GitHub3.9 KiB
skill.oms.sig(opens on GitHub)Not mirrored: opens on GitHub7.8 KiB

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Needs: Needs Python

Security

What our scanner noticed· Medium risk

Large base64-encoded blobs
Scripts that your agent may run

This skill includes scripts that your agent may run. Read them before you use it.

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Diagnosing Bugs
Matt Pocock· Community
A structured, six-phase discipline for systematically diagnosing and resolving complex software bugs and performance regressions.
Software developmentClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository317.5k
Systematic Debugging
Jesse Vincent (obra)· Community
A rigorous, four-phase framework for identifying and resolving technical issues by prioritizing root cause analysis over symptomatic fixes.
Software developmentClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository314k
Performance Optimization
Addy Osmani· Community
A rigorous, measurement-first workflow for identifying and fixing performance bottlenecks in frontend and backend applications.
Software developmentClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository113.9k
Debugging And Error Recovery
Addy Osmani· Community
A systematic, six-step framework for triaging, reproducing, and resolving software bugs to ensure root-cause fixes.
Software developmentClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository113.9k
Animation Performance
Emil Kowalski· Community
A diagnostic guide for identifying and fixing frame drops in web animations to ensure smooth, performant motion.
Software developmentClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository54.3k
Parallel Debugging
Seth Hobson· Community
A structured framework for debugging complex issues using parallel hypothesis testing and evidence-based arbitration.
Software developmentClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository42.3k