Artificial Intelligence
Supervisely MCP connector
The Supervisely connector brings your annotation objects, tags, and labeling agents into Claude, ChatGPT, Cursor, and any other MCP client, so your team can review, tag, merge, and clean up labels while every action runs inside their own Supervisely access.
-
How it connects. Connects with an API key. The credential goes into a vault nobody reads back.
-
One address. https://api.elaichi.ai/mcp, the same for every user.
-
Their own access. An agent never gets more than the person it acts for.
How to connect
How to connect Supervisely to Claude, ChatGPT or Cursor
Two steps, about a minute.
In Elaichi
Connect Supervisely once
-
Open Connections, choose Add connection, and pick Supervisely.
-
Optionally set Share with to give a team access, then press Connect.
-
Paste a Supervisely API key. One person generates a token in Supervisely and pastes it once. Everyone else works through Share with, and never sees it.
The credential is vaulted. Nobody reads it back, not even the AI.
Add connection
Choose a connector.
In your AI client
Point it at one endpoint
Everyone in the organization uses the same address, and each person only ever reaches what their own account allows.
Connect Supervisely to Claude
-
1
Open Customize, then Connectors.
-
2
Press Add.
-
3
Name it, paste the MCP server URL, then Continue.
https://api.elaichi.ai/mcp -
4
Sign in and approve.
On Team and Enterprise, an Owner adds it once. Everyone else turns it on for themselves.
Connect Supervisely to ChatGPT
-
1
Open Plugins, then press the + button.
-
2
Name it and paste the endpoint into Server URL.
https://api.elaichi.ai/mcp -
3
Leave Authentication on OAuth, then tick the risk acknowledgement.
-
4
Press Create, then sign in and approve.
Works on the web today. The plugin directory lives at chatgpt.com/plugins.
Connect Supervisely to Cursor
-
1
Open
~/.cursor/mcp.json. -
2
Add the endpoint under
mcpServers.https://api.elaichi.ai/mcp -
3
Reload Cursor, then sign in and approve.
~/.cursor/mcp.json
{
"mcpServers": {
"elaichi": {
"url": "https://api.elaichi.ai/mcp"
}
}
}
Set up per machine, so repeat it on each computer you work from.
Connect Supervisely to any MCP client
-
1
Add the endpoint as a remote MCP server.
https://api.elaichi.ai/mcp -
2
Sign in and approve.
{
"mcpServers": {
"elaichi": {
"url": "https://api.elaichi.ai/mcp"
}
}
}
The Elaichi Agent already has these tools, with nothing to set up.
Use cases
What teams do with Supervisely through Elaichi
Every one of these runs inside the access the person already has, and lands in the same audit log.
-
Annotation
Tag a batch of objects in one go
Ask for every object in a set to get a class or attribute tag applied at once, instead of clicking through them one by one.
-
Annotation
Merge or split labels after review
Ask to merge duplicate objects that were drawn twice, or split an object that should have been two, without opening the editor.
-
Machine learning
Check which agents are online
Ask which Supervisely agents are available and ready before kicking off a training or inference job.
-
Machine learning
Upgrade agents across the fleet
Ask to upgrade the agents that are behind and set their status, so nothing runs on an outdated version.
-
Quality review
Find objects with missing tag values
Ask for the tag values on a group of annotation objects and spot the ones that were left blank or set inconsistently.
-
Operations
Clean out bad labels in bulk
Ask to remove a list of annotation objects that failed review, once the removal has been approved and the objects are confirmed.
Try asking
- “List Supervisely agents that are available right now.”
- “Which projects gained the most annotation objects this week?”
- “Add the review tag to yesterday's labeled objects.”
AI tools
Supervisely tools for your AI agents
455 tools are ready to call through Elaichi's MCP endpoint the moment you connect Supervisely, governed by the same roles, restrictions, and audit log as everything else in Elaichi.
No tools match your search.
See it in Elaichi
What connecting Supervisely gets you
6 screens from the product, each doing one job for your Supervisely account.
The agent
Ask about agents and annotations, get answers.
Plain language questions run against live Supervisely agents, projects and annotation tags.
- agents
- projects
- annotation objects
- tags
Ask Elaichi to work across your apps.
List Supervisely agents that are available right now.
Which projects gained the most annotation objects this week?
Add the review tag to yesterday's labeled objects.
Also runs in Claude, ChatGPT or Cursor
MCP clients
Claude, ChatGPT and Cursor share one endpoint.
Connect any MCP client to Supervisely over OAuth, no SDK, no shared key.
Copy the endpoint
Tool catalog
455 Supervisely tools ready without custom code.
Browse tool names, resources and methods across agents, projects, annotations and tags.
- Create a Supervisely agents net set status
- Supervisely agents exists
- Supervisely agents available
- List all Supervisely agents
Toolboxes
Every team gets its own Supervisely toolbox.
Curate one toolbox per team so labeling, ML and ops see different tools.
- Labeling
- ML engineering
- Data operations
- Quality review
Shared connections
Teammates use Supervisely without seeing a credential.
The connections table shows who connected each account and how widely it is shared.
- Labeling EMEA
- ML engineering
- Data operations
- Quality review
Audit log
Every Supervisely call is written down.
When, who, what happened, type and resource, in an append only log.
- When
- Who
- What happened
- Type
Launching soon
From answering questions to doing the work
A person no longer has to ask. A trigger starts the work, inside the same permissions and the same audit log as everything else. Automations and live dashboards are launching soon, on the Black plan.
Automations
A schedule fires, a digest ships.
Six steps fetch Supervisely jobs, group them, draft a digest, await approval, post back.
Supervisely digest
Run 418 · started 2 minutes ago · on behalf of Emma Laurent
-
✓
Schedule
Every weekday at 08:00
0.2s -
✓
Fetch agents
Supervisely
1.4s -
✓
Group by owner
Transform
0.1s -
✓
Draft the digest
Agent step
Ran with 4 tools, returned a structured summary
6.2s -
Approve the digest
Needs approval
Assigned to Michael Brennan
Approve Deny -
Post the digest
Supervisely
Queued
Collections and dashboards
Supervisely numbers arrive before anyone asks.
Four metrics, 14 days of created annotations and a team breakdown, refreshed on schedule.
Supervisely health
Refreshed 4 minutes ago · every 15 minutes · from the agents collection
Agents
1,284 ↓ 12%
Projects
96 ↓ 8%
Needs attention
3 ↑ 2
Updated this week
412 ↑ 9%
Agents created
Last 14 days
By team
Share of activity
Labeling EMEA 34%
ML engineering 27%
Data operations 21%
Quality review 18%
FAQ
Frequently asked questions
How do I connect Supervisely to Claude?
Two steps. First, connect Supervisely in Elaichi, which asks you to paste your Supervisely API key; there is no OAuth application to register and no client ID or secret to generate. Then in Claude go to Customize, then Connectors, then Add, and paste the endpoint https://api.elaichi.ai/mcp. Sign in with your Elaichi account and Supervisely is ready to use in Claude.
Does Supervisely work with ChatGPT and Cursor as well as Claude?
Yes. Once Supervisely is connected in Elaichi, the same endpoint, https://api.elaichi.ai/mcp, works in Claude, ChatGPT, Cursor, any other MCP client, and the Elaichi Agent. You connect Supervisely once and use it from every client.
What can an AI agent actually do with my Supervisely data?
With Supervisely connected, an agent can list your annotation objects, look up a single object, add tags to one object or to a whole batch, merge, split, or swap objects, and update or remove objects in bulk. It can also list your Supervisely agents, check which are available, create new ones, upgrade them, and change their status. Because Supervisely offers hundreds of actions, short concrete asks like "tag these objects as vehicle" work better than long sentences.
Does connecting Supervisely give the AI access to my whole workspace?
No. Every call to Supervisely runs inside the access of the person who signed in, so an agent sees the same teams, projects, and annotation objects that person can already reach in Supervisely. Elaichi can narrow that access further with restrictions, and it can never widen it beyond what the person already has.
Can I stop an agent from deleting or changing things in Supervisely?
Yes. Restrictions in Elaichi work per action, so you can allow reading annotation objects and adding tags while blocking bulk removal or object updates in Supervisely. A blocked action is never advertised to the AI client at all, so no prompt can reach it.
What happens to a Supervisely connection when someone leaves?
Offboarding a person in Elaichi ends their access to Supervisely through every client at once. A shared Supervisely connection keeps working for everyone else on the team. If you want to remove Supervisely entirely, disconnecting it once in Elaichi removes it from Claude, ChatGPT, Cursor, and every other client.
Put Supervisely in front of your team
Fourteen days on Gold, no credit card. Connect it once and pick what each team can call.