- name
SKILL.md
技能檔案保留在發布者的儲存庫中。
在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add artokun/comfyui-mcp --skill krea2-txt2img -a cursor -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add artokun/comfyui-mcp --skill krea2-txt2img -a claude-code -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add artokun/comfyui-mcp --skill krea2-txt2img -a codex -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add artokun/comfyui-mcp --skill krea2-txt2img -a grok -y在 Grok Bot 聊天中傳送複製的訊息。Bot 會從 GitHub 取得這個 SKILL.md,並加入你所有 Bot 共用的技能中。
Install this skill: https://github.com/artokun/comfyui-mcp/tree/HEAD/plugin/skills/krea2-txt2img來自發布者
Krea 2 is a 12B-parameter Diffusion Transformer from Krea.ai (released June 2026, weights open-sourced under the Krea 2 Community License, free commercial use up to 50 seats). Two variants:
Sliced from the KREA2 ULTRA V2 monolith into standalone single-pipeline packs. Pick by how you prompt and what you want:
krea2-txt2img-manual: plain prose prompt (the MANUAL PROMPT node).krea2-txt2img-json: Ideogram-4-style structured JSON / area prompting
(Ideogram4PromptBuilderKJ).krea2-combo: two-pass detail boost, a first pass then a low-denoise
refine (denoise 0.3), with the krea2 turbo LoRA @0.2 on both passes plus
the optional IdeoKrea LoRA. JSON/Ideogram-style prompting; saves both
passes to compare.Each pack's one prompt source is active (no prompt-mode bypass to flip).
ImageSharpenKJ runs before SaveImage. V2 adds the Krea2T-Enhancer MODEL
detail-boost patch (ships active) and drops v1's ConditioningKrea2Rebalance.
RBG_Smart_Seed_Variance ships bypassed (optional, see below).
Krea 2 has native ComfyUI support (comfy/text_encoders/krea2.py, ComfyUI ≥
v0.26.0). The CLIPLoader uses type=krea2, with a Qwen3-VL 4B text
encoder and the Qwen image VAE. The Qwen3-VL encoder drives strong prompt
adherence and structured-JSON prompts.
Aitrepreneur/FLX mirror; official: krea/Krea-2-Turbo)UNETLoader (krea2_turbo) → CLIPLoader (type=krea2) → VAELoader
(qwen_image_vae), wired via KJNodes SetNode/GetNode buses into a subgraph
(CLIPTextEncode → KSampler → VAEDecode). An rgthree Any Switch sits in
front of the encoder; in each pack only that pack's prompt source is wired to it
(manual node in -manual, JSON builder in -json).Ideogram4PromptBuilderKJ, ImageSharpenKJ, INTConstant.capitan01R): Krea2T-Enhancer, the V2 MODEL→MODEL
detail-boost patch, wired in the model path (PowerLora → Krea2T-Enhancer →
sampler). Ships active; bypass to compare against the un-boosted result.RBG_Smart_Seed_Variance, optional, ships
bypassed in the positive-conditioning loop.cubiq): ImageResize+, combo only (the two-pass
VAE-roundtrip resize).er_sde, scheduler simple are the verified defaults.Krea2T-Enhancer) + comboKrea2T-Enhancer is a MODEL→MODEL patch (the V2 "massive detail boost"). It
sits inline in the model path and ships active in all three packs. Widgets
are [on, strength, …]; bypass it (or toggle on) to A/B the boost.krea2-combo is the full demonstration of the boost, a two-pass refine:
FIRST PASS (8 steps, er_sde, denoise 1) → VAE roundtrip → SECOND PASS (4 steps,
euler, denoise 0.3), with the turbo LoRA @0.2 on both passes. It SAVES
BOTH passes so you can see the boost. The IdeoKrea LoRA is downloaded but NOT
wired by default. Drop it into the Power Lora Loader's empty slot (start ~0.5 to
1.0; it's a test LoRA) for the turbo + IdeoKrea Ideogram-style combo.All packs leave RBG_Smart_Seed_Variance in the positive-conditioning loop
bypassed (passthrough). Un-bypass on the live canvas with panel_set_node_mode
(or in the UI) for controlled variations of the same prompt without changing the
composition. Set its seed mode to randomize and tune the variance mode (e.g.
🌿 Balanced) / strength widgets. Leave bypassed for a deterministic result.
Like Ideogram 4, Krea 2's Qwen3-VL encoder reads structured prompts (per-region
desc + bounding boxes + palettes). For structured prompting use the
krea2-txt2img-json pack; its Ideogram4PromptBuilderKJ drives the encoder
directly (no bypass to flip). After the render, VERIFY the image matches the JSON
you set (view it) BEFORE continuing; if it doesn't, a field is probably stale.
Fix and rerun. Gotchas learned the hard way:
background,
technical, style, lighting (widgets 3/5/6/7). Leaving stale values leaks
content (a leftover celebrity portrait bled into a tea still-life).palette: []
(top-level and per-box) gives a clean full-frame result.style for object/landscape
scenes. Krea 2 follows it well.-manual / -json core graph): render-verified, crisp 1920×1080 / 8
steps / cfg 1 / er_sde (snow-leopard prose + tea-still-life JSON with each object
in its bbox).Krea2T-Enhancer active patch + the krea2-combo two-pass)
are statically validated (clean slice + structural lint) but not yet
live-rendered. They need the ComfyUI-Krea2T-Enhancer node, the turbo/IdeoKrea
LoRAs installed, and a healthy ComfyUI. Re-run scripts/verify-render.mjs once
those are present.ImageSharpenKJ (rcas 0.55) before SaveImage is active.
Bypassing it drops the image link (a converter gap: bypass-passthrough doesn't
cross a subgraph IMAGE output), and the contrast-adaptive sharpen suits Krea 2's
crisp look anyway.CLIPLoader: 'krea2' not in list → ComfyUI too old; update to ≥ v0.26.0.Torch not compiled with CUDA enabled → reinstall torch for your CUDA tag
(--index-url https://download.pytorch.org/whl/cu128).packs/ and observed renders; not a vendor prompting guide.Build Krea 2 Turbo txt2img workflows with the native krea2 CLIPLoader, Qwen3-VL encoder, Qwen image VAE, 8-step turbo settings, and Ideogram-style JSON prompting
---name: krea2-txt2imgdescription: Build Krea 2 Turbo txt2img workflows with the native krea2 CLIPLoader, Qwen3-VL encoder, Qwen image VAE…---# plugin/skills/krea2-txt2img/SKILL.md常見問題
從發布者下載 plugin/skills/krea2-txt2img/SKILL.md,或複製路徑後開啟儲存庫。PlugMCP 只收錄這項技能,不代管檔案。
發布者將它存放在 https://github.com/artokun/comfyui-mcp。
Claude、Cursor、Codex 和 Grok Build 可以讀取你從發布者儲存庫加入的 SKILL.md。在聊天中傳送技能的 GitHub 資料夾連結,Grok Bot 就會安裝它。