triton-sageattention

triton-sageattention is a skill for Claude Code from sandyup/comfyui-mcp. It costs 170 tokens per session (5,610 once invoked), scanned C, a copy of triton-sageattention, MIT.

Optional GPU software for ComfyUI, an image and video generation interface. Triton compiles GPU code, while SageAttention changes how model attention is calculated to speed up supported video workflows.

In plain words
What is it for?
Installing Triton and SageAttention in the correct ComfyUI Python environment on Windows or Linux. It supports SageAttention modes and Triton-based PyTorch compilation in workflows such as WanVideoWrapper.
Why use it?
It provides the components that some ComfyUI video workflows expect and can reduce generation time. Installation must match the exact Python, PyTorch and CUDA versions, or the setup may fail or produce errors.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions Codex.

Needs its repository: it reads a path above its own folder, which exists only inside the repository. The line is See also [`comfyui-launch-flags`](../comfyui-launch-flags/SKILL.md) for the full.

Part of the comfy plugin — 32 skills, 11 commands, 4 agents, 2 hooks shipped together

Good fit Installing Triton and SageAttention in the correct ComfyUI Python environment on Windows or Linux. It supports SageAttention modes and Triton-based PyTorch compilation in workflows such as WanVideoWrapper.

Compare 6 skills from other repositories ↓
Install

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.

Clone the repo
git clone --depth 1 https://github.com/sandyup/comfyui-mcp
agentmods
npx agentmods add skills/sandyup/comfyui-mcp/triton-sageattention

Made for: Claude Code.

Or install comfy, the plugin that ships this one along with the rest of its 32 skills, 11 commands, 4 agents, 2 hooks.

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 triton-sageattention

README.md
[![agentmods](https://agentmods.dev/badge/skills/sandyup/comfyui-mcp/triton-sageattention/github.svg)](https://agentmods.dev/skills/sandyup/comfyui-mcp/triton-sageattention)
Your own site
<a href="https://agentmods.dev/skills/sandyup/comfyui-mcp/triton-sageattention"><img src="https://agentmods.dev/badge/skills/sandyup/comfyui-mcp/triton-sageattention/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 triton-sageattention

Your own site · 80×15
<a href="https://agentmods.dev/skills/sandyup/comfyui-mcp/triton-sageattention"><img src="https://agentmods.dev/badge/skills/sandyup/comfyui-mcp/triton-sageattention.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 170 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,610 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin 86% copy Near-identical to another mod 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 $0.00170 $0.05610
Opus 5 $0.00085 $0.02805
Sonnet 5 $0.00034 $0.01122
Haiku 4.5 $0.00017 $0.00561

Measured 10d ago against content hash e199b6170b9a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade C, and why

triton-sageattention scanned grade C with 1 finding 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

rm -rf ~/.triton
Origin

This is a copy

86% identical to triton-sageattention — 270 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

plugin/skills/triton-sageattention/SKILL.md · 384 lines

How it starts

The opening of the file, as written. The whole thing — 384 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Triton + SageAttention (ComfyUI acceleration)

See also comfyui-launch-flags for the full attention / VRAM / cache flag matrix. Note the Z-Image exception: Z-Image is broken under --use-sage-attention → launch it with --use-pytorch-cross-attention instead.

Overview

Two optional accelerators that many modern video graphs (especially kijai's ComfyUI-WanVideoWrapper) reference by default:

  • SageAttention (import sageattention) — a quantized attention kernel. Selected via a node's attention_mode = sageattn (WanVideoWrapper) or ComfyUI's --use-sage-attention startup flag. ~20–40% faster sampling on supported NVIDIA GPUs.
  • Triton — the GPU kernel compiler that inductor torch.compile needs. WanVideoWrapper's WanVideoTorchCompileSettings (and any torch.compile/ inductor node) compiles the model through Triton for another speedup.

⚠️ The risk. Both are version-locked to your exact torch + CUDA + python. A wrong wheel doesn't just fail to install — it can break the torch install (mismatched CUDA DLLs, ImportError, or silent NaNs). And the failure mode of not having them is a hard crash before any sampling: ValueError: Can't import SageAttention: No module named 'sageattention', or compile errors / triton: unavailable in the startup log. This is exactly the video-extend TRAP 5.

Therefore the default is: get a working render FIRST with the sdpa / no-compile fallback, then OFFER to install acceleration for speed. Never silently run a torch-breaking install to "fix" a workflow — fall back, render, then ask.

⚠️ Verification note (June 2026). Wheel sources, the triton↔torch table, and the live attention_mode enum below were verified against woct0rdho/triton-windows, woct0rdho/SageAttention releases, and WanVideoWrapper's nodes (see Sources). Versions move fast — always re-read the live torch/CUDA/python first (commands below) and pick the wheel that matches; flag anything you can't confirm rather than guessing.

Read the full file on GitHub · 384 lines

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 · 384 lines · 0 tokens per session scan C e199b6170b9a

Subscribe to this mod's changes

triton-sageattention is a skill published in the GitHub repository sandyup/comfyui-mcp (1 stars, last pushed 2mo ago), licensed MIT. It adds 170 tokens to every session and 5,610 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). It is 86% identical to triton-sageattention, differing in 270 lines, and is treated as a copy.

Related

Other skills, from other repositories

webgl-holographic-foil

A self-contained WebGL2 hero: thin-film interference over a crushed-foil surface whose palette shifts with the viewing angle; move the cursor to tilt the film.

nexu-io/open-design · 41 tokens

general-video

Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated…

heygen-com/hyperframes · 92 tokens

html-ppt-hermes-cyber-terminal

OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.

nexu-io/open-design · 53 tokens

html-ppt-taste-brutalist

16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).

nexu-io/open-design · 78 tokens

diagnostic-stem-delivery

Audio production with diagnostic analysis, timecode parsing from documents, and verified export workflow.

HKUDS/OpenSpace · 23 tokens

chengfeng-check-updates

An environment manager for a video-editing system. It checks whether its skills and runtime—the software needed to run them—are installed and compatible.

Agentchengfeng/chengfeng-videocut-skills · 120 tokens