Langfuse MCP connector
The Langfuse connector brings your datasets, annotation queues, comments and review assignments into Claude, ChatGPT, Cursor, Microsoft Copilot and the Elaichi Agent, so your team can ask about them, add to them and tidy them in plain language.
- Connects with an API key
- Free 14-day trial
- No credit card
List open annotation queue items assigned to the eval team
Working in Langfuse with Elaichi
-
List all Langfuse public annotation queues
-
Get single Langfuse public annotation queue by ID
-
List all Langfuse annotation queue items
How to connect Langfuse to Claude, ChatGPT, Cursor or Microsoft Copilot
Step by step, for each AI app
Langfuse MCP connector for 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.
Langfuse MCP connector for Claude Code
-
1
In a terminal, run
claude mcp addwith the endpoint.https://api.elaichi.ai/mcp -
2
Start Claude Code and run
/mcp. -
3
Pick elaichi, then sign in and approve in the browser.
Terminal
$ claude mcp add --transport http elaichi --scope user https://api.elaichi.ai/mcp
Use --scope project instead to share it with the repository through .mcp.json.
Langfuse MCP connector for 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. The plugin directory lives at chatgpt.com/plugins.
Langfuse MCP connector for 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.
Langfuse MCP connector for Microsoft Copilot
-
1
In Copilot Studio, open your agent, then Tools, Add a tool, New tool and Model Context Protocol.
-
2
Paste the endpoint into Server URL, choose OAuth 2.0, then Dynamic discovery, and press Create.
https://api.elaichi.ai/mcp -
3
Create a new connection, sign in and approve, then press Add to agent.
-
4
Under Channels, open Teams and Microsoft Copilot and make the agent available in Microsoft Copilot.
- Server URL
- https://api.elaichi.ai/mcp
- Authentication
- OAuth 2.0
- Type
- Dynamic discovery
Microsoft Copilot was called Microsoft 365 Copilot. Keep the tool on end-user credentials, so each person signs in as themselves.
Connect Langfuse 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.
What teams do with Langfuse through Elaichi
Ask in plain English. Each person’s AI can do only what they’re allowed to, and every action it takes in Langfuse is logged.
-
AI engineering
Turn a real conversation into a test case
After spotting a bad answer in production, add it to a Langfuse dataset as a new item with the expected output, without leaving the chat where you found it.
-
Quality
Queue outputs for human review
Create a Langfuse annotation queue for this week's release, add the traces that need eyes on them, and assign each reviewer their share.
-
Product
Check what is waiting for review
Ask how many items are still open in a Langfuse annotation queue and who they are assigned to before the weekly quality meeting.
-
Data
Record how an evaluation run went
Log dataset run items in Langfuse as a batch of test cases finishes, so every run is tied back to the dataset it came from.
-
Research
Leave notes on traces for the team
Add a comment to a Langfuse trace or observation explaining why the output was wrong, so the next person reading it has context.
-
Operations
Clean up finished review work
Mark reviewed Langfuse queue items as done, remove stale assignments in bulk, and retire dataset items that no longer reflect the product.
Try asking
- “List open annotation queue items assigned to the eval team”
- “Create an annotation queue for last week's flagged traces”
- “Which blob storage integrations are configured in Langfuse?”
What connecting Langfuse gets you
6 screens from Elaichi, each doing one job for your Langfuse account.
Ask Elaichi to work across your apps.
List open annotation queue items assigned to the eval team
Create an annotation queue for last week's flagged traces
Which blob storage integrations are configured in Langfuse?
Also runs in Claude, ChatGPT, Cursor or Microsoft Copilot
The agent
Ask about annotation queues in plain language.
Elaichi's agent answers from live Langfuse queues, items and assignments.
Copy the endpoint
AI apps
Claude, ChatGPT, Cursor and Microsoft Copilot reach Langfuse.
One org MCP endpoint over OAuth, no SDK and no shared API key.
Tool catalog
100 Langfuse tools ready to call.
Queues, queue items, assignments and blob storages, with no custom code written.
Toolboxes
Every team gets its own Langfuse toolbox.
Curate one toolbox for evaluation, another for platform, each scoped to that team.
Shared connections
Teammates use Langfuse without seeing a key.
The connections table shows who connected each Langfuse account and its sharing reach.
Audit log
Answer who touched which queue item.
Append-only log of when, who, what happened, type and Langfuse resource.
From answering questions to doing the work
A trigger starts the work, inside the same permissions and the same audit log as everything else. Automations, collections and live dashboards come with the Black plan.
Langfuse digest
Run 418 · started 2 minutes ago · on behalf of Emily Carter
-
✓
Schedule
Every weekday at 8:00 AM
0.2s -
✓
Fetch annotation queues
Langfuse
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 Jake Morgan
Approve Deny -
Post the digest
Langfuse
Queued
Automations
A schedule drafts your Langfuse digest.
Six steps fetch queue items, group them, draft, wait for approval, then post back.
Langfuse health
Refreshed 4 minutes ago · every 15 minutes · from the annotation queues collection
Annotation queues
1,284 ↓ 12%
Queue items
96 ↓ 8%
Needs attention
3 ↑ 2
Updated this week
412 ↑ 9%
Annotation queues created
Last 14 days
By team
Share of activity
Eval Prod 34%
ML Engineering 27%
Platform 21%
Data Science 18%
Collections and dashboards
Langfuse queue health, computed on a schedule.
Four metrics, 14 days of items created, and a breakdown by team.
For IT and engineering
Elaichi vs Zapier MCP vs Composio for Langfuse
All three can connect Langfuse to an AI assistant, and all three have admin controls. They differ in where access lives and how you pay.
| What to check | Elaichi | Zapier MCP | Composio |
|---|---|---|---|
| Where the AI connects |
One address for the whole organization.
https://api.elaichi.ai/mcp
|
A server per member, created at sign-in. | An MCP endpoint per team, or an SDK. |
| Control over Langfuse tools | Allow or restrict single Langfuse tools, per role or user. | App and action restrictions on the account. | Role permissions, down to the action. |
| Record of calls | One audit entry per Langfuse call. | A History tab of tool calls. | A log of every tool call. |
| Single sign-on | SAML or OIDC, plus SCIM, on Gold. | SAML on Enterprise. | SAML and OIDC on Enterprise. |
| Price | Free for one person; $15 per user per month on Gold. | 2 tasks per successful call. | Billed per tool call. |
Sources: Zapier MCP docs, security, usage; Composio docs, gateway, enterprise, pricing. Checked September 2026.
Longer take: Zapier MCP alternative and when you don't need an MCP gateway.
Langfuse tools for your AI agents
100 tools are ready to call through Elaichi's MCP endpoint the moment you connect Langfuse, governed by the same roles, restrictions, and audit log as everything else in Elaichi.
No tools match your search.
Questions, answered
How do I connect Langfuse to Claude?
Connect Langfuse in Elaichi by pasting an API key from your Langfuse project settings. 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 https://api.elaichi.ai/mcp. Sign in with your Elaichi account and Langfuse is available in the chat.
Does Langfuse work with ChatGPT and Cursor as well as Claude?
Yes. Once Langfuse is connected in Elaichi, the same endpoint, https://api.elaichi.ai/mcp, works in Claude, ChatGPT, Cursor, Microsoft Copilot, any other MCP client and the Elaichi Agent. You connect Langfuse once and every client picks it up.
What can an AI agent actually do with my Langfuse data?
With Langfuse connected, an agent can list your datasets and pull up individual dataset items, add new items and dataset run items, create annotation queues, add items to them and assign reviewers, and read or leave comments on traces. Because Langfuse exposes a lot of actions, the agent looks them up as needed, so short concrete asks like "add this to the regression dataset" work better than long paragraphs.
Does connecting Langfuse give the AI access to every project?
No. The connection runs inside whatever the Langfuse API key you pasted can already see, typically one project, and the agent gets nothing beyond that. Elaichi can narrow access further, for example read only on datasets, but it can never widen what Langfuse itself allows.
Can I stop an agent from deleting or changing things in Langfuse?
Yes. Restrictions in Elaichi work per action, so you can allow reading Langfuse datasets and queues while restricting deletes of dataset items, queue items or assignments. A restricted action still appears in Elaichi's tool search, flagged as restricted, but Claude, ChatGPT, Cursor, Microsoft Copilot or any other client cannot run it, whatever the prompt says.
What happens to a Langfuse connection when someone leaves?
When a person is offboarded in Elaichi, their access to Langfuse through every client ends at once. A Langfuse connection they shared keeps working for everyone else on the team. If you want to remove Langfuse entirely, disconnecting it once in Elaichi removes it from Claude, ChatGPT, Cursor, Microsoft Copilot and the Elaichi Agent together.
Does the Langfuse MCP connector work with Gemini, Codex, Claude Code or other MCP clients?
Yes. Langfuse is reached over the same MCP endpoint every client uses, so anything that speaks MCP can call it — Gemini, Codex, Claude Code, Windsurf, Cline, Zed and OpenCode among them — alongside Claude, ChatGPT, Cursor, Microsoft Copilot, and the Elaichi Agent. The tools on offer and the access behind them are identical whichever client asks. Only the setup screen differs.
Is Elaichi an alternative to Zapier MCP for Langfuse?
Yes. Both let Claude, ChatGPT or Cursor use Langfuse. Zapier MCP fits a team that already automates in Zapier, since each person signs in and acts as themselves in that account. Elaichi fits when IT wants one address for the whole company, per-tool rules by role, and a record of every Langfuse call.
How is Elaichi different from Composio for Langfuse?
Composio gives AI agents tools and sign-in handling across 1,000+ apps, for developers building agents or people using an assistant, billed per tool call. Elaichi gives a company's own people governed access to Langfuse: one address, restrictions per role or user, and $15 per user per month. Both have role permissions and a log of every call.
Put Langfuse in front of your team
14 days of Black, no credit card. Connect it once and pick what each team can call.
When the trial ends, nothing is deleted. You move to the free Personal plan, and connections, roles and the audit log stay where they are.
Add one link to your AI app
The same link for everyone on your team.
https://api.elaichi.ai/mcp
ChatGPT
Claude
Cursor
Microsoft Copilot