SkillAgentSearch skills...

Comfiest

A local-first, open-weights production suite for ComfyUI. Your agent runs image, video, speech, and music workflows on your own GPU.

Install / Use

claude mcp add BuffMcBigHuge -- npx -y github:BuffMcBigHuge/Comfiest

If the server publishes to npm under a different name, use that package instead β€” check the repo README.

About this skill
πŸ”Œ

MCP Server

Model Context Protocol server

Quality Score

84/100

Category

Automation

Supported Platforms

Claude Code
Claude Desktop

Our assessment of Comfiest

Comfiest scores 84/100 on our quality scale, 1991st of 2,758 Automation skills we index.

Its MCP Server is 24 KB long, well organised into 19 sections with 5 code examples: a thorough specification that gives an agent plenty to work with.

It has 10 GitHub stars, so there is little community track record yet; judge it on its content.

Substance
30/30
Structure
20/20
Description
15/15
Adoption
4/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated yesterday, so Comfiest is actively maintained.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 97/100, with no cautions. These come from repository metadata, not a code audit β€” read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands.

Automated pattern scan on 2026-11-21. It catches known dangerous patterns, not every risk β€” read a skill before letting an agent act on it.

Comfiest compared with similar skills

All 4 of these similar skills score higher than Comfiest; compare them before choosing.

SkillScoreStarsUpdatedFormat
Comfiest (this skill)by BuffMcBigHuge84101d agoMCP Server
taste-skillby Leonxlnx100120.0ktodayCLAUDE.md
Agent-Reachby Panniantong100114.0ktodayCLAUDE.md
hyperframesby heygen-com10088.2ktodayCLAUDE.md
rufloby ruvnet10082.3ktodayCLAUDE.md

Frequently asked questions

How do I install Comfiest?
Run claude mcp add BuffMcBigHuge -- npx -y github:BuffMcBigHuge/Comfiest. The install tabs above show the steps for each supported agent.
Which AI agents does Comfiest work with?
It is written for Claude Code and Claude Desktop, as a MCP Server file. Other agents that read the same format can often use it too.
Is Comfiest safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. It is MIT-licensed and scores 97/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is Comfiest still maintained?
The repository was last updated yesterday, so Comfiest is actively maintained.
<p align="center"> <a href="https://comfie.st"> <img src="assets/logo.png" alt="Comfiest β€” a sleeping cat on a cozy purple armchair" width="480"> </a> </p> <h1 align="center">Comfiest</h1> <p align="center"> <strong>comΒ·fiΒ·est</strong> <em>adjective</em><br> <em>informal</em> : the superlative form of <strong>comfy</strong> β€”<br> the most comfortable, cozy, or providing the greatest degree of physical ease and relaxation.<br> <strong>comfy</strong> <em>noun</em> : ComfyUI </p> <p align="center"> <strong>A local-first, open-weights video production suite for ComfyUI β€” driven by your agent.</strong> </p> <p align="center"> <a href="https://comfie.st"><img src="assets/icon.png" alt="" width="20" height="20" align="absmiddle"> <strong>Comfie.st</strong></a> Β· built by <a href="https://byMAR.CO">byMAR.CO</a> </p> <p align="center"> <a href="https://comfie.st">Website</a> Β· <a href="#start-here-point-your-agent-at-the-repo">Start here</a> Β· <a href="#quick-start">Quick start</a> Β· <a href="#workflow-catalog">Workflows</a> Β· <a href="#requirements">Requirements</a> Β· <a href="#contributing">Contributing</a> </p> <p align="center"> <a href="https://youtu.be/MOtF6weKl6c"> <img src="assets/video-thumb.jpg" alt="Watch the video: Comfiest" width="640"> </a> </p>

New in Comfiest

NEW β€” MiniMax H3 just leveled up. Comfiest ships the latest H3 innovations as runnable graphs and agent skills, not just links:

  • Pika H3 Turbo LoRAs β€” the official Pika H3 Flex adapters from Pika-Labs/Pika-H3-Flex at the 6-step LCM contract, 24 fps, with text, first/last-frame, and reference modes. Pika T2V
  • Upscale refiner LoRA β€” the Regenerate-2K refiner LoRA plus latent 2Γ— upscale for 2560Γ—1440 finals. Upscale workflows
  • Fidelity 2K refiner β€” a 0.2 GB rank-21 Open-H3-Regenerate-2K-IC LoRA plus LMS Sharp, with no HyperFlow and no patching. It holds 32.4 dB PSNR to a Pika draft and keeps the draft audio bit-aligned. IC upscale
  • 16 MP native image model β€” MiniMax-H3-Image: 33.1B, 6-step CFG 1, 4–16 MP natively, exact multi-line text, and 1/2/4-reference edit graphs (up to 16 supported). H3-Image T2I, edit, 2-ref, 4-ref
  • Streaming 4Γ— video upscale β€” ReCaVSR: one-step diffusion VSR on Wan2.2, 16.9 MP on 24 GB, with source audio and frame rate preserved. ReCaVSR X4
  • Recast or edit any clip β€” Viggle Remake: a two-LoRA H3 pair for character swaps, object removal and weather/setting changes, driven by painted frames; source motion, camera and soundtrack are kept. Swap, edit
  • Looping videos β€” seamless 154-frame H3 loops with synchronized audio and a Γ—2 seam check, built on the MiniMax H3 base. Perfect loop

These are brand-new innovations in the MiniMax H3 ecosystem, and Comfiest supports all seven.

Comfiest packages the parts of an AI production pipeline that normally live in someone's head: model routing, workflow templates, hardware limits, submission loops, output provenance, and the small traps that waste a render. Point a coding agent at this repository and it can set up the MCP bridge, pick the right model for a shot, run it, and collect the result.

Built for the Comfy Org Open Source Higgsfield challenge.

| Create | Models and capabilities | | --- | --- | | Images from text | Krea 2 Turbo, Flux.2 Klein 9B, MOSS-Image, MiniMax-H3-Image, P-H3-Image | | Image edits | Krea 2 Edit, Flux.2 Klein 9B reference edit (1–4 references), MiniMax-H3-Image (1/2/4-ref graphs; up to 16 supported), P-H3-Image (up to 9 references) | | Speech | Fish Audio Drama 3 β€” single/dual speaker, voice clone | | Music | YuE2 3B, MiniMax Music 3 | | Video from text or references | MiniMax H3 β€” Pika H3 Flex 6-step default, fastvideo variants, 1–2 pictures, video/audio references, forced lip-sync, character swap | | Video recast / edit | Viggle Remake β€” H3 LoRA pair for character swaps, object removal, and setting/weather changes | | Video with joint audio | LTX-2.5 22B distilled β€” text, identity reference, or first frame | | Video upscale / restoration | RelayVSR X2 β€” one-step pixel-space 2Γ—; ReCaVSR X4 β€” one-step streaming 4Γ— with audio passthrough | | 2K upscale / refine | MiniMax H3 refiner β€” latent-space t2v/v2v upscale to 2560Γ—1440, with 2-step re-synthesis or 3-step IC fidelity | | Seamless loops | MiniMax H3 Perfect Loop β€” text-to-looping-video with synchronized audio | | Mattes / segmentation | SAM3 β€” text-prompted image/video luma mattes, native ComfyUI nodes |

No partner nodes. No API keys. No credits. Everything runs on your own ComfyUI with open weights.


Start here: point your agent at the repo

The primary interface is an agent. Open this repository in Claude Code, OpenCode, Codex, Cursor, or any client with MCP support, and say:

β€œSet up Comfiest for my ComfyUI server, then run the fox demo and show me the output.”

The agent should:

  1. Read README.md, COMFY-SETUP.md, and skills/SKILL.md.
  2. If ComfyUI is not installed, follow COMFY-SETUP.md to install ComfyUI, the node packs, and the models.
  3. Install/register the comfy-* MCP server for your ComfyUI host using MCP-SETUP.md.
  4. Confirm the target with server_info() and /system_stats.
  5. Pick a workflow from the routing table and submit it with run_workflow.
  6. Collect the output into work/outputs/<project>/<date>/<prompt_id>/ and verify its size.

Add the skills to your agent

If your agent supports skills:

  • Copy skills/SKILL.md to <client-skills>/comfiest/SKILL.md for the umbrella rules and routing.
  • Copy skills/minimaxh3/, skills/minimaxh3-pika/, skills/minimaxh3-character-replace/, skills/minimaxh3-upscale/, skills/minimaxh3-perfect-loop/, skills/ltx2-5/, skills/relayvsr/, skills/ph3-image/, skills/h3-image/, skills/recavsr/, skills/viggle-remake/, skills/sam3/, skills/prompt-writing/, skills/workflow-development/, and skills/fish-audio-drama3/ as sibling skill directories if your client discovers each skill separately. skills/prompt-writing/ contains eight family skills; copy its child directories individually if your client requires a flat layout.

The umbrella carries the hard rules and routing; the subskills carry family and recipe detail.

What your agent gets

  • Routing that respects hardware. Krea/Flux/MOSS/H3-Image for images, YuE2/Music 3 for songs, Drama 3 for speech, MiniMax H3/Pika for video, 2K refine, 4Γ— upscale, recast/edit, and seamless loops. Heavy models are called out before they blow up a 24 GB card.
  • Workflow templates that actually run. UI files for ComfyUI, pre-slimmed api/*.api.json mirrors for MCP execution, and conversion notes for the frontend-only nodes that break API runs.
  • A development loop for new workflows. Scratch copies, UI-to-API conversion, target preflight, submit/poll/collect, run-log entries, and a promotion checklist in skills/workflow-development/SKILL.md.
  • A native matte primitive. SAM3 image/video segmentation with open-vocabulary prompts, coverage checks, and alpha-WebM compositing β€” no character-replacement graph required.
  • A second video family. LTX-2.5 22B distilled generates video and audio in one pass from text, an identity reference, or a first frame.
  • Prompt-writing skills for every family. Krea, Flux, MOSS, H3-Image, LTX, YuE2, Music 3, and Drama 3 prompts get the model-specific structure instead of a one-line caption.
  • A submission loop that survives real life. wait: false + /history polling, direct output URL fallback when local job state lags, one-job-per-target discipline.
  • Provenance by default. Every file lands under a prompt_id folder with a run-log entry; vision QA is allowed only on your own outputs.
  • A writable workspace. work/ is the only place agents create files; the shipped workflows stay intact.

Quick start

1. Set up ComfyUI

Install ComfyUI 0.45 or newer with the node packs and models listed in skills/SKILL.md. A 24 GB-class GPU is the baseline. Drama 3 loads 38.6 GB of experts into system RAM; use a high-RAM host.

For a complete, copy-pasteable install β€” ComfyUI, all 18 custom-node packs, every model, and the TensorRT depth engine β€” follow COMFY-SETUP.md. It includes two scripts:

python scripts/install_nodes.py --comfyui "$COMFYUI"
python scripts/download_models.py --comfyui "$COMFYUI"

2. Connect the MCP bridge

Comfiest talks to ComfyUI through comfy-mcp, a thin stdio server around comfy-cli.

# macOS / Linux
python -m venv ~/venvs/comfiest
~/venvs/comfiest/bin/pip install comfy-cli comfy-mcp
# Windows PowerShell
python -m venv $env:USERPROFILE\venvs\comfiest
& "$env:USERPROFILE\venvs\comfiest\Scripts\pip" install comfy-cli comfy-mcp

Then register one server per ComfyUI target. Full instructions, environment variables, and client snippets for OpenCode, Claude Code/Desktop, Codex, Cursor, Hermes, and a generic stdio client are in MCP-SETUP.md.

3. Run your first workflow

Ask the agent for the fox demo, or submit directly:

const run = await tools["comfy-<host>-<port>"].run_workflow({
  workflow_path: "<repo>/workflows/text2image-krea.json",
  wait: false
});
await tools["comfy-<host>-<port>"].job({ action: "wait", prompt_id: run.prompt_id });
await tools["comfy-<host>-<port>"].fetch_outputs({
  prompt_id: run.prompt_id,
  out_dir: "<repo>/work/outputs/fox-demo/<date>/" + run.prompt_id
});

Replace <repo> with the absolute path to this repository on the MCP host, and <host>/<port> with one of your registered servers.

Demo inputs

The demo prompts are self-contained, but the reference-driven workflows expect files on the ComfyUI host:

| Workflow | Expected input | How to supply it | | --- | --- | --- | | Krea / Flux edit | example.png | Stage in work/uploads/<project>/, upload with upload_file. The Klein 2/3/4-ref graphs need every LoadImage slot set; all defaults are example.png. | | Drama 3 clone variants | example.wav | Upload clean single-speaker clips and replace the transcript placeholders to match. | | MiniMax H3 reference variants | example.png, example.mp4, example.wav | Upload the slots the variant uses; every image slot uses the same placeholder. | | LTX-2.5 ref2v / i2v | example.png | Upload a subject image for ref2v, or a first frame for i2v; replace the file in work/uploads/<project>/. | | RelayVSR | example.mp4 | Upload the source clip; output is 2Γ— its size and keeps its audio. | | ReCaVSR | example.mp4 | Upload the sourc

Truncated for display β€” read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars10
CategoryAutomation
Updated1d ago
Forks2

Languages

Python

Trust signals

97/100

From repository metadata: license, adoption, age and documentation. Not a code audit β€” see the Safety scan above for what the skill file itself contains.

1 info
Comfiest β€” MCP Server: Install & Safety Check | SkillAgent