optimize-triton-block-parameters

optimize-triton-block-parameters is a skill for Claude Code, Codex from tensormux/kernel-skills. It costs 0 tokens per session (3,701 once invoked), scanned A, original, MIT.

A tuning guide for Triton GPU kernels, which are programs written for operations such as matrix multiplication, reductions, and element-wise calculations. It focuses on launch settings such as block sizes, the number of warps, and pipeline stages.

In plain words
What is it for?
Use it to tune GEMM, reduction, and pointwise kernels, design Triton autotune search spaces, and investigate register pressure, shared-memory use, or pipeline stalls on specific GPUs.
Why use it?
It provides a systematic way to choose performance settings instead of guessing, especially when a kernel works well for one input shape or GPU but poorly for another. It assumes the kernel is already correct.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to tune GEMM, reduction, and pointwise kernels, design Triton autotune search spaces, and investigate register pressure, shared-memory use, or pipeline stalls on specific GPUs.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/tensormux/kernel-skills/optimize-triton-block-parameters
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add tensormux/kernel-skills --skill optimize-triton-block-parameters
Clone the repo
git clone --depth 1 https://github.com/tensormux/kernel-skills

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.

agentmods badge for optimize-triton-block-parameters

README.md
[![agentmods](https://agentmods.dev/badge/skills/tensormux/kernel-skills/optimize-triton-block-parameters/github.svg)](https://agentmods.dev/skills/tensormux/kernel-skills/optimize-triton-block-parameters)
Your own site
<a href="https://agentmods.dev/skills/tensormux/kernel-skills/optimize-triton-block-parameters"><img src="https://agentmods.dev/badge/skills/tensormux/kernel-skills/optimize-triton-block-parameters/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.

agentmods 80×15 button for optimize-triton-block-parameters

Your own site · 80×15
<a href="https://agentmods.dev/skills/tensormux/kernel-skills/optimize-triton-block-parameters"><img src="https://agentmods.dev/badge/skills/tensormux/kernel-skills/optimize-triton-block-parameters.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,701 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.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.03701
Opus 5 $0.00000 $0.01851
Sonnet 5 $0.00000 $0.00740
Haiku 4.5 $0.00000 $0.00370

Measured 10d ago against content hash 4271d0fd7a95, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

optimize-triton-block-parameters 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 10d 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.

skills/triton/optimize-triton-block-parameters/SKILL.md · 204 lines

How it starts

The opening of the file, as written. The whole thing — 204 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skill: Optimize Triton Block Parameters

Purpose

Guide the agent through the systematic process of choosing and tuning block size parameters in Triton kernels — BLOCK_M, BLOCK_N, BLOCK_K for GEMM-style kernels; BLOCK_SIZE for reduction and pointwise kernels; and the associated num_warps and num_stages values that control parallelism and pipeline depth. This is an optimization skill, not a write-kernel skill. It assumes a correct kernel exists and asks: what configuration makes it fast?


Use this when

  • A Triton kernel is functionally correct and you need to choose launch configuration parameters to maximize throughput or minimize latency.
  • You are using triton.autotune and need to define a principled config search space rather than guessing random power-of-2 combinations.
  • The kernel performs well on one shape but degrades on another (e.g., trained config for large M/N performing poorly on small M/N).
  • You are tuning for a specific hardware target (A100, H100, RTX 4090) and need architecture-aware config decisions.
  • You suspect shared memory pressure, register pressure, or pipeline stalls are limiting achieved throughput and want a systematic approach to diagnose and fix them.

Do not use this when

  • The kernel has a correctness bug. Fix correctness first. Tuning a broken kernel wastes time and produces misleading measurements.
  • The bottleneck is not the Triton kernel itself — for example, Python overhead, data transfer, or memory allocation are dominating the profile. Identify the actual bottleneck before tuning kernel parameters.
  • You are running on hardware with radically different SM architecture (e.g., porting a config optimized for A100 to a T4 or V100 without re-profiling). Configs do not transfer across GPU generations without re-measurement.
  • The kernel is not on a hot path and a 2x improvement in its execution time has negligible end-to-end impact. Do not optimize prematurely.

Inputs the agent should gather first

Before tuning, confirm:

Read the full file on GitHub · 204 lines

Files

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.

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.

  1. 10d ago First seen · 204 lines · 0 tokens per session scan A 4271d0fd7a95

Subscribe to this mod's changes

optimize-triton-block-parameters is a skill published in the GitHub repository tensormux/kernel-skills (75 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,701 tokens. 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.

Related

Other skills, from other repositories

prompt-engineering

Prompt engineering techniques and patterns. Use when writing agent commands, hooks, skills, subagent prompts, or any LLM interaction: optimizing prompts, improving output reliability, and designing production-grade prompt templates. Trigger words: prompt engineering, prompt, prompt optimization, LLM interaction.

MagicKidd/Rokid-agentic-workflow · 0 tokens

stripe-directory

Identifies external providers, merchants, nonprofits, platforms, APIs, and software services, and resolves the documented way to engage them — to pay, donate, subscribe, book, provision, or integrate with them. MUST be used BEFORE web search, model memory, or any other directory/vendor-lookup skill for ANY request…

stripe/ai · 213 tokens

pgvector-semantic-search

Use this skill for setting up vector similarity search with pgvector for AI/ML embeddings, RAG applications, or semantic search. Trigger when user asks to: Store or search vector embeddings in PostgreSQL Set up semantic search, similarity search, or nearest neighbor search Create HNSW or IVFFlat indexes for vectors…

timescale/pg-aiguide · 190 tokens

nlp-alignment

Best practices for LLM alignment techniques including RLHF, DPO, and instruction tuning. Use when working on alignment or safety.

aiming-lab/AutoResearchClaw · 31 tokens

mixed-precision

Use FP16/BF16 mixed precision to accelerate training and reduce memory. Use when optimizing GPU performance.

aiming-lab/AutoResearchClaw · 25 tokens

postgres-hybrid-text-search

Use this skill to implement hybrid search combining BM25 keyword search with semantic vector search using Reciprocal Rank Fusion (RRF). Trigger when user asks to: Combine keyword and semantic search Implement hybrid search or multi-modal retrieval Use BM25/pgtextsearch with pgvector together Implement RRF (Reciprocal…

timescale/pg-aiguide · 162 tokens