Dragoneye

Computer vision models with no training required.

CommunityOnlineNo sign-inGlobalFreeCan modify data

What it can do

  • Create Model: Create a new Dragoneye custom computer-vision model to detect objects, their categories and attributes. Creation is asynchronous: the call returns the model in PENDING status with its uu
  • List Models: List the calling user's custom models, newest first, paginated by cursor. Returns up to `limit` models (default 50, max 200) plus a `next_cursor` to fetch the following page. `next_cursor
  • Get Model: Fetch a single custom model by uuid, including its current status, full model_state, and (for FAILED/REVERTED models) a structured error. Use this to poll the outcome of create_model / upda

What data it sees

Do you need an account

No: the server works without sign-in

Computer vision models with no training required. Create and manage custom zero-shot vision models that can do object detection, category classification and attribute classification.

How to connect

How to connect Dragoneye to Claude

The server address to paste into Claude settings:

https://mcp-server--dragoneye.run.tools

  1. Open Claude (claude.ai or the desktop app).
  2. Go to Settings → Connectors.
  3. Click Add custom connector.
  4. Paste the server address copied below into Remote MCP server URL and click Add.
  5. In a chat, click + → Connectors and switch the new connector on.

Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

Authorization

After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

How to connect Dragoneye to ChatGPT

The server address to paste into ChatGPT settings:

https://mcp-server--dragoneye.run.tools

  1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
  2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
  3. Open Settings → Connectors and click Create.
  4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
  5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
  6. In a new chat open + → Apps/Connectors and enable the connector.

OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

Authorization

On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

How to connect Dragoneye to Cursor

The server address to paste into Cursor settings:

https://mcp-server--dragoneye.run.tools

Fastest: click Open in Cursor below and confirm the prompt.

Manually:

  1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
  2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
  3. Save the file. The server appears in the MCP list; authorize it there if asked.

Authorization

If the service needs sign-in, an authorize button appears next to the server in the MCP list.

Server tool list (6)

Raw names from tools/list. Only developers need these.

create_modelCreate a new Dragoneye custom computer-vision model to detect objects, their categories and attributes. Creation is asynchronous: the call returns the model in PENDING status with its uuid. Poll get_model with that uuid until status is COMPLETED (success) or REVERTED (with an error). Typical poll interval: a few seconds. The `name` is unique per billing account — collisions return MODEL_NAME_CONFLICT.
list_modelsList the calling user's custom models, newest first, paginated by cursor. Returns up to `limit` models (default 50, max 200) plus a `next_cursor` to fetch the following page. `next_cursor` is null on the last page. Treat the cursor as opaque — do not parse it.
get_modelFetch a single custom model by uuid, including its current status, full model_state, and (for FAILED/REVERTED models) a structured error. Use this to poll the outcome of create_model / update_model. Status transitions are PENDING → IN_PROGRESS → COMPLETED, or → FAILED/REVERTED.
update_modelUpdate a custom model's name and/or model_state. Re-runs model generation asynchronously when the state changes; returns PENDING and the same polling rules apply as for create_model. It can update a model in COMPLETED and REVERTED state. Refuses to update a model that already has an in-progress update (PENDING or RUNNING) (MODEL_UPDATE_IN_PROGRESS) or sits in FAILED state (MODEL_IN_FAILED_STATE) — for FAILED, use delete_model + create_model to start fresh. When ONLY `name` changes (state byte-identical), uses a synchronous rename fast-path that skips re-generation.
rename_modelRename a custom model. Synchronous: no re-generation, no PENDING state. It can update a model in COMPLETED and REVERTED state. Refuses if the model is currently being updated (MODEL_UPDATE_IN_PROGRESS) or in FAILED state (MODEL_IN_FAILED_STATE). Renaming to the current name is a no-op.
delete_modelPermanently delete a custom model. Cannot be undone. Refuses to delete a model with an in-progress update (MODEL_HAS_ACTIVE_UPDATE) — wait for it to settle, then retry. Models in FAILED state CAN be deleted and this is the supported way to recover from FAILED.