Langfuse MCP Server — Governed AI Access | Barndoor
Langfuse MCP Server: Governed AI Access
Langfuse's MCP server includes 50 different actions available to any connected model or AI agent. When registered with Barndoor, you can ensure centralized governance, visibility, and control across every AI agent your team uses.
Tools
50
Capabilities
Category
Data, Analytics & Code Quality
Type
Official Remote
What can AI agents do with the Langfuse MCP server?
Tools Available
| Action | Type | Description |
|---|---|---|
| getPrompt | read | Fetch a specific prompt by name with optional label or version parameter. |
| listPrompts | read | List and filter prompts in the project. |
| createTextPrompt | create | Create a new text prompt version in Langfuse. |
| createScore | create | Create one score in the current Langfuse project. |
| queryMetrics | read | Answer analytics questions about usage, model costs, latency, errors, scores, or grouped breakdowns. |
| deleteScoreConfig | delete | Delete a score configuration from the current Langfuse project by archiving it. |
Use cases for Langfuse
- Prompt Management: Create and version text and chat prompts, Fetch prompts with resolved dependencies, Update prompt labels and versions
- LLM Observability: List and inspect traces and observations, Filter observations by custom criteria, Track generations, spans, and tool calls
- Evaluation and Scoring: Create and manage score configurations, Run dataset experiments and evaluations, Set up annotation queues for review
- Usage Analytics: Query metrics for cost and latency, Analyze LLM usage across models, Track errors and performance trends
What are the risks of using the Langfuse MCP without an AI gateway?
- Uncontrolled reads: AI agents can pull data out of Langfuse with no visibility into what was accessed.
- Ungoverned writes: agents can create or modify records in Langfuse with no approval step.
- Destructive actions: agents can delete or overwrite Langfuse data with no approval gate or audit trail.
- Native OAuth is all-or-nothing: connecting directly grants an agent everything your user account can do in Langfuse, with no per-tool scoping.
Why connect Langfuse with the Barndoor AI gateway?
Barndoor MCP Governance is the gateway between your AI agents and the tools they use.
| At a glance | Available? |
|---|---|
| Fine-grained access controls policies per agent, team, role, user, MCP server and tool call | Yes |
| Approval required before write/destructive actions | Yes (configurable per role) |
| Full audit log (reads and writes) | Yes |
| Sensitive data detection before it reaches a model or agent | Yes |
| Instant access revocation | Yes |
| SOC 2 Type II compliant | Yes |
- Centralized MCP management — One place to manage and control every MCP server across the organization so your team has an approved catalog of MCP servers that employees and AI agents connect through.
- Universal connection — One universal URL works across every AI provider your team uses.
- Fine-grained access controls — Set access policies for how AI is connecting with Langfuse based on human identities, agent profiles, tool-access, context, and data fields.
- Protect sensitive data — Keeps sensitive data from reaching a model or AI agent with PII detection, masking, tokenization, redaction, and more.
- Observability — Real-time reporting and dashboards that give you visibility into what AI is doing with Langfuse, how policies are being enforced, and who (human or agent) is doing what.
- Enterprise-grade security — Stay compliant with audit trails and logging.
Works with every AI client
Getting started with Langfuse in Barndoor
- Connect the Langfuse MCP server once
- Define your policies
- Go live across every AI client
Frequently asked questions
What is MCP?
MCP (Model Context Protocol) is the open standard that lets AI models and agents connect to external tools and data sources through a common interface, instead of a custom integration for every tool. Langfuse is an MCP server that exposes a defined set of tools an AI agent can call. Barndoor sits at that connection point, applying access policies and audit logging to every tool call regardless of which AI client is making it.
Is Barndoor secure?
Yes. Barndoor is SOC 2 Type II compliant. Barndoor enforces fine-grained access control, requires approval before destructive actions, protects sensitive data with data loss prevention, and logs every action for audit.
How are Langfuse schema changes handled?
Barndoor tracks tool and schema changes on Langfuse. New or changed tools inherit deny-by-default access until an admin reviews and approves them, keeping runtime enforcement current without manual reconfiguration. Every schema change is captured in the audit trail alongside who approved it.
Do you integrate with my identity provider?
Barndoor authenticates admins and policy managers through your OIDC or SAML 2.0 identity provider. Signing into Claude, Cursor, or ChatGPT still happens directly with those providers. Barndoor's integration point is the Langfuse connection where access policies apply to what an agent can do once it's connected.
Do users have to enter tokens manually?
No. Barndoor brokers the OAuth connection to Langfuse centrally, so users and agents never handle or store a standing token themselves. Access is granted and revoked through policy instead of shared credentials.
How does Barndoor integrate with AI agent identities?
Barndoor gives each AI agent its own governed identity called an AgentProfile, separate from the human who configured or uses it. AgentProfile lets policies and audit logs attribute a given action to the specific agent, not a shared service token, so you can set different access rules per agent and revoke one agent's access without affecting others.
Can I deploy custom-built MCP servers?
Yes. Barndoor supports customer-built MCPs, so internal tools and systems your team builds get the same policy enforcement and audit trail as Barndoor MCPs and official MCPs.
Can I set and audit tool-level access across teams?
Yes. Barndoor's access control is fine-grained, down to the agent, team, group, user, MCP server, the specific tool call, and the data type involved. Every call is logged with a timestamp, the user or agent calling it, the client it came from, and the result of the policy applied, so you can see exactly who or what did what, when, and whether it was allowed, across any team or role.
How can I get started with Langfuse and Barndoor?
Connect Langfuse once, define your access policies, and go live across every AI client your team uses. Setup takes three steps: connect the server, define policies by role or agent, and roll out to Claude, Cursor, ChatGPT, and any other MCP-compatible client. Request a demo at barndoor.ai/get-a-demo to learn more.