Generate AI animationswithClaude XToonkit

Connect Toonkit to Claude
Toonkit MCP

What can you do with Toonkit MCP?

Learn how to connect and use Toonkit MCP in ChatGPT and Claude.

Claude

 

Toonkit

Image nodeVideo nodeNode connection

I've connected the character, background, and video nodes. Open each node on the canvas to review and adjust them.

Ready to create

Preparing the chat example.

Start a conversation and the results will appear here.

TOONKIT MCP SKILLS

Among so many AI video MCPs, why Toonkit MCP?

From picking models and connecting references to token-aware lookups and cost checks.

1

Enough with just “make it”

Skill_Crafted Workflows

Even a short “make it” request needs order and judgment to produce. We built in human-refined production know-how so the AI carries on from setup through generation to checking results.

2

Because your credits matter

Skill_Cost-Aware Execution

It checks the cost and sets a cap before running. Even if the response drops, it recovers the same request to prevent duplicate generations and charges.

3

Because tokens matter as much as credits

Skill_Context Efficiency

It reads production rules once per session and looks up only the model details it needs. We refined even how the AI fetches information so it doesn't reread full lists unnecessarily.

4

Because your time choosing models matters

Skill_Model Selection

Focus on the result you want. With a selection guide covering each model's strengths, supported features, and cost, the AI picks the right model for your request.

5

Because your imagination should come through as is

Skill_Reference Direction

We carefully refined how the AI connects and uses references so the intent in your references and prompts carries through generation.

6

Because we think about what comes after the result

Skill_Connected Creation

It connects and organizes references, generation settings, and finished results on one Canvas. Your next edit can build on existing assets and workflows so you can keep creating.

01Canvas Skills

Skills that let the AI agent create, connect, and edit Canvas nodes itself.

canvas_list_modelsList Canvas image, video, and voice models and their supported settingscreate_canvasCreate an empty Canvaslist_canvasesList my Canvasesget_canvasGet a Canvas with its nodes, connections, generation results, and pending changesrename_canvasRename a Canvasdelete_canvasSoft-delete a Canvas. Generations already run and their files are keptcanvas_add_text_nodeAdd a text nodecanvas_update_text_nodeEdit a text node's contentcanvas_add_media_nodeAdd existing images, videos, or audio as Canvas nodescanvas_add_generation_nodeAdd a node with generation settings. No run, no chargecanvas_update_generation_nodeEdit a generation node's prompt and settingscanvas_connectConnect nodes. Connecting a reference input inserts a reference chip into the promptcanvas_disconnectRemove a node connectioncanvas_group_nodesGroup related nodes under a name and arrange them by flowcanvas_delete_nodesDelete selected nodes and their connections. Generation history is keptcanvas_generate_imageGenerate an image on the Canvascanvas_generate_videoGenerate a video on the Canvas. Supports image, video, and audio references depending on the modelcanvas_generate_voiceGenerate a voice on the Canvaslist_voicesList available voicesget_canvas_mutationCheck whether a Canvas change request was appliedestimate_creditsEstimate the credits a Canvas generation needsget_credit_balanceGet your credit balance and daily usage limit

023D Reference

Skills that let the AI agent compose a 3D scene, place objects, and set the framing.

catalogcanvas_reference3d_catalog · List supported objects, joints, motion presets, edit commands, and limitsget_scenecanvas_reference3d_get_scene · Get a 3D node's scene and the revision to edit againstcreatecanvas_reference3d_create · Request a 3D node with an empty scene or a person + camera templateeditcanvas_reference3d_edit · Request a batch of scene edit commands. Rejected if the revision is out of dateexportcanvas_reference3d_export · Request a 3D scene export as an image or video. Uses no credits

03Playground

Skills that generate images, videos, and voices directly through the AI agent.

list_modelsplayground_list_models · List Playground models and their supported settingsestimate_creditsplayground_estimate_credits · Estimate generation costgenerate_imageGenerate an image from text or reference imagesgenerate_videoGenerate a video from text, first/last frames, or reference mediaretry_generationRetry a failed generation with the same settings. Charged again

04Asset Skills

Skills that let the AI agent browse, upload, and manage global assets.

list_assetsList your library and browse by type, keyword, or folderget_mediaGet media details and access URLscreate_uploadIssue a URL for uploading an imageconfirm_uploadConfirm the upload and register the media for references

05Guide Skills

Skills that let the AI agent check Toonkit's production guide and look up job status.

get_generation_guideGet the generation rules and usage guide provided by the serverlist_generationsList MCP generation jobs with their status and billingget_generationGet a job's status, results, and inputs, and refresh result URLs
FAQ

Frequently asked questions

Questions? Contact us at cs@toonkit.io.

What is Toonkit MCP?

A tool that connects AI agents to Toonkit's creation features so you can use them through conversation. MCP is a standard that lets AI use external services' tools — describe what you want, and the AI runs Toonkit's features for you.

Which AI tools can I use it with?

Use it with ChatGPT and Claude. Check the connection guide above for each tool's support and setup.

What can I create with Toonkit?

Connect the Toonkit MCP/plugin and describe what you want in conversation, from image, video, voice, and music generation to building Canvas-based projects.

Do I need to know how to draw or code?

Instead of drawing or coding, you can start by describing what you want in words. Describing details like your character's features, expressions, background, and composition makes it easier to convey the direction.

Do I need an API key to connect?

No separate API key is needed. Follow the guide above to enter the commands in your terminal, then just allow access to connect right away.

How much does it cost?

Toonkit MCP uses the existing Toonkit credit system as-is. There's no separate MCP fee, and the credits you hold work the same in connected agents.

How long does a task take?

Task time can vary with the type of work and generation options. You can keep checking progress in your terminal, and ask it anything anytime!

Can I keep using my existing characters or images?

You can continue your next task from existing images or previous results. Describe the traits you want to keep and the parts you want to change.

A new way to make AI animation,
Toonkit MCP.

Connect Toonkit MCP now and create your own AI animations with just prompts.

Back to top
ToonkitToonkit

© 2026 Toonkit. All rights reserved.

Innerverz Co., Ltd. | CEO : Junghyuk Im | Business Registration : 615-88-01959 | Mail-order Sales Registration : 2026-Seoul Gangnam-05795

Address : 4F, 19, Nambusunhwan-ro 351-gil, Gangnam-gu, Seoul, Korea | Tel : 070-4102-3345 | CS : cs@toonkit.io