MCP integration

MCP is a connection method that lets you plug the 데브다이브-모두의창업 AI API directly into AI tools like Claude and Cursor. Once connected, you can call features like listing models, generating text, analyzing images, and running workflows right inside the AI tool you already use.

No separate installation or server setup is needed. You just register the connection address and API key.

---

What you need to connect

1. Connection address (MCP endpoint) — https://modoo.devdive.me/mcp
2. API key — the sk-modoo-... key you issued from the dashboard

---

Registering it with a tool

Claude, Cursor, Claude Code, and similar tools register the connection with a JSON configuration like the one below. Enter this into each tool's MCP (or connector) settings screen.

{
  "mcpServers": {
    "modoo": {
      "url": "https://modoo.devdive.me/mcp",
      "headers": {
        "Authorization": "Bearer sk-modoo-..."
      }
    }
  }
}

• url — the connection address above
• Authorization in headers — put your issued API key in the form Bearer sk-modoo-...

Confirming the connection

After registering, restart the AI tool and the modoo tools will appear in the list. When you make a natural-language request like "tell me the available models," the tool is called automatically.

---

Available tools

Once connected, you can use the following tools inside your AI tool.

Tool	What it does	Input
list_models	List the available models	(none)
get_credits	Check remaining credits	(none)
get_usage	View recent usage history	limit (optional)
get_job	Check an async job's status/result	job_id
generate_text	Generate text	prompt, model (optional), system_prompt (optional)
text_to_speech	Speech synthesis (TTS)	text, model (optional: modoo-tts-minimax|modoo-tts-eleven), voice_id (optional; required for ElevenLabs), voice_model (optional), ElevenLabs options stability·similarity_boost·style·speed·use_speaker_boost·language_code·output_format·previous_text·next_text·seed·apply_text_normalization (optional)
transcribe	Speech to text (STT)	audio_url or encoded_audio, language (optional)
analyze_image	Analyze an image	image_url or encoded_image
analyze_document	Analyze a document	file_url or encoded_file, operation (optional)
generate_image	Generate an image from text	prompt, model/size/n (optional)
edit_image	Edit an existing image, keeping its content (image-to-image)	prompt, one of image_url/image_urls/encoded_image, model (optional)
generate_video	Generate a video from text	prompt, model (optional: modoo-video|modoo-video-happyhorse-t2v), negative_prompt (optional), HappyHorse options resolution·ratio·duration·watermark·seed (optional)
animate_image	Turn an existing image into a video (image-to-video)	prompt, image_url (https address), end_image_url·driving_audio_url·first_clip_url (optional, modoo-video-i2v only — closing-frame / driving-audio / clip-continuation https addresses), model (optional: modoo-video-i2v|modoo-video-happyhorse-i2v), negative_prompt (optional), HappyHorse options resolution·duration·watermark·seed (optional)
reference_to_video	Generate a video from reference images (reference-to-video)	prompt, reference_image_urls (list of 1–9 https addresses), model (optional, default modoo-video-happyhorse-r2v), resolution·ratio·duration·watermark·seed (optional)
generate_music	Generate music (turn lyrics into a song)	lyrics, model (optional, default music-2.5), prompt (optional)
web_search	Web search (Brave)	query
run_workflow	Run a workflow (pass blocks directly, executed sequentially)	blocks
run_workflow_by_id	Run a workflow saved in the devdive editor by its id	workflow_id
get_workflow_run	Fetch a persisted workflow run (status, per-block results, charge)	run_id

Input forms: image/audio/document tools take either a URL (image_url/audio_url/file_url) or encoded data (encoded_*, the file converted into text form). Most AI tools handle this automatically when you drop in a file.

Async jobs: generate_image, edit_image, generate_video, animate_image, reference_to_video, generate_music, and transcribe take time, so the tool waits until completion and returns the final result (image/video/music URL, text). If it runs very long and times out, you get a job_id back — check again later with get_job.

Workflows: run_workflow executes the blocks in order, chaining an earlier block's output into a later one via {{block_id.field}}. Runs are persisted — call get_workflow_run with the returned run_id to look them up again anytime. For a workflow built in the devdive editor, skip writing blocks and pass just the id from the editor URL (/workspace/editor/{id}) to run_workflow_by_id.

For how to write the blocks of a workflow (run_workflow), see the Workflows documentation.

---

Security note

Handle your API key carefully

The Authorization header in the connection settings contains your API key as-is. This key is like a password, so don't share it with others or post it in a public place (for example, a public repository). If you think your key has been exposed, reissue it immediately from the dashboard.

---

What's next

• Detailed spec for each feature → API Reference
• Chaining several tasks together → Workflows
• Error code explanations → Error guide