Langfuse Integration

Ingest and query LLM traces, manage prompts and datasets, and run Langfuse evaluations and experiments. Connect Langfuse to the rest of your stack so AI outputs can trigger real workflow actions. Start with Get Annotation Queue, then use Get Annotation Queue Item or Get Blob Storage Integration to turn results into useful next steps.

Explore Documentation

3-day trialCancel anytimeUp to 10 minutes quick set up

GDPR
Data Encryption
2FA
Local Data Region

Supported triggers and actions for Langfuse integration

Triggers can detect changes in Langfuse and run actions that can create, search or modify data inside Langfuse automatically [3] — exactly how you set them.

Get Annotation QueueReturns annotation queue.
Get Annotation Queue ItemReturns annotation queue item.
Get Blob Storage IntegrationReturns blob storage integration.
Get CommentReturns comment.
Get DatasetReturns dataset.
Get Dataset ItemReturns dataset item.
Get Dataset RunReturns dataset run.
Get Evaluation RuleReturns evaluation rule.
Get EvaluatorReturns evaluator.
Get Evaluator VersionsReturns evaluator versions.
Get MediaReturns media.
Get MetricsReturns metrics.
Get ProjectsReturns projects.
Get PromptReturns prompt.
Get Prompt VersionReturns a specific prompt version (explicit reuse of the Get Prompt operation, pinned to a version number).
Get ScoreReturns a score (supports both trace and session scores).
Get SessionReturns session.
Get TraceReturns trace.
List Annotation Queue ItemsRetrieves a list of annotation queue items.
List Annotation QueuesRetrieves a list of annotation queues.
List Blob Storage IntegrationsRetrieves a list of blob storage integrations.
List CommentsRetrieves a list of comments.
List Dataset ItemsRetrieves a list of dataset items.
List Dataset RunsRetrieves a list of dataset runs.
List DatasetsRetrieves a list of datasets.
List Evaluation RulesRetrieves a list of evaluation rules.
List EvaluatorsRetrieves a list of evaluators.
List Experiment ItemsRetrieves a list of experiment items.
List ExperimentsRetrieves a list of experiments.
List LLM ConnectionsReturns llm connections.
List ObservationsRetrieves a list of observations.
List PromptsRetrieves a list of prompts.
List ScoresRetrieves a list of scores.
List SessionsRetrieves a list of sessions.
List TracesRetrieves a list of traces.
Add Annotation Queue ItemCreates annotation queue item.
Create Annotation QueueCreates annotation queue.
Create CommentCreates comment.
Create DatasetCreates dataset.
Create Dataset ItemCreates dataset item.
Create Evaluation RuleCreates evaluation rule.
Create EvaluatorCreates evaluator.
Create PromptCreates prompt.
Create ScoreCreates score.
Create TraceCreates trace (submitted via the legacy batch ingestion endpoint — Langfuse removed the direct v1 POST /traces operation).
Delete Annotation Queue AssignmentsDeletes annotation queue assignments.
Delete Blob Storage IntegrationDeletes blob storage integration.
Delete Dataset ItemDeletes dataset item.
Delete Dataset RunDeletes dataset run.
Delete Evaluation RuleDeletes evaluation rule.
Delete EvaluatorDeletes evaluator.
Delete LLM ConnectionDeletes llm connection.
Delete PromptDeletes prompt versions. If neither version nor label is specified, all versions of the prompt are deleted.
Delete ScoreDeletes score.
Delete TraceDeletes trace.
Ingest Traces (OpenTelemetry)Performs ingest traces (opentelemetry).
Make an API CallPerforms an arbitrary authorized API call.
Submit FeedbackPerforms submit feedback.
Update Annotation Queue ItemUpdates annotation queue item.
Update Blob Storage IntegrationUpdates blob storage integration.
Update Evaluation RuleUpdates evaluation rule.
Update EvaluatorUpdates evaluator.
Update LLM ConnectionUpdates llm connection.
Update Prompt Version LabelsUpdates labels for a specific prompt version.

How to integrate Langfuse in automation workflow?

3 simple steps to make your Langfuse connected with other apps in automation workflows on WEXTL. Up to 10 minutes integration set up.

STEP 1

Select Langfuse in workflow

Pick Langfuse app in the workflow builder [4] - just one click to select the app.

STEP 2

Connect Langfuse

When prompted, authorize access to Langfuse in WEXTL once. Langfuse credentials are encrypted and never shared with WEXTL. Authorization will be reused in any future API calls [3].

WEXTL wants to integrate with your Langfuse account. Approve?

STEP 3

Build a workflow

Build your workflow with Langfuse yourself or with WEXTL® AI Builder. Connect Langfuse to any available integrations [5], or use HTTP node to go beyond them.

Langfuse

How Langfuse connects and integrates with WEXTL

Save and approve your credentials once — WEXTL® stores authorization information encrypted and reuses the authentication whenever Langfuse app is called in workflows. That allows your ai automations to run without your attention, per defined schedule [7]. Explore documentation [6] for a detailed list of fields that Langfuse supports.

Full Langfuse API documentation

The following table illustrates what fields Langfuse integration require during authorization flow inside our workflow builder [4].

Demonstration of Langfuse connection step inside WEXTL Workflow Builder [4].

Field	Type	Help
HostingRequired	select	Langfuse Cloud region or your self-hosted deployment.
Base URL	url	Self-hosted API origin including /api/public — e.g. https://langfuse.example.com/api/public.
Public KeyRequired	password	—
Secret KeyRequired	password	—

Connect Langfuse to popular apps on WEXTL

Enter a name of the other app you want to sync with Langfuse below. Click on selected app to explore integration options with Langfuse. If the app you search for is not available in WEXTL yet, you can use HTTP node [10] instead to integrate it directly.

Example automation with Langfuse integration

Select Get Annotation Queue action from Langfuse - this will be a starting point in your automation. Whenever event happens in Langfuse, workflow will start [7] from Get Annotation Queue node.

Search for Add Annotation Queue Item, connect it as the next step of integration sequence. You can use filters between Get Annotation Queue and Add Annotation Queue Item, so only important data pass from Get Annotation Queue to the next step. WEXTL gives you full control over the process of Langfuse automation. All parts of workflow are visible on the canvas [4].

Langfuse on WEXTL has 64 modules for combining ai automations. Explore Langfuse documentation for more.

Langfuse

Get Annotation Queue

DeepL

Performs action

DeepL

Triggers

Langfuse

Add Annotation Queue Item

Demonstration of how visual Langfuse automation [6] looks in WEXTL.

Frequently asked questions

Common questions about automating Langfuse in your workflows.

Langfuse has no triggers, but it provides 64 actions including "Get Annotation Queue", "Get Annotation Queue Item", and "Get Blob Storage Integration". Pair them with triggers from other apps to automate Langfuse tasks.

Integration has 64 actions, including "Get Annotation Queue", "Get Annotation Queue Item", and "Get Blob Storage Integration", and more. Refer to our Langfuse Documentation for detailed inputs and outputs.

No. Most automations are built visually on the canvas by picking triggers, actions, and mapping fields. You can add custom JavaScript with Functions when advanced logic is needed — but our customers do that in rare cases [4].

Yes. Connections are authenticated per account, credentials are encrypted at rest, and you can enforce approvals and team-level access controls on any automation.

The Langfuse integration on WEXTL runs directly on its vendor public API — every field name, authentication method, and rate limit is controlled by Langfuse official API and covered in documentation [3]. WEXTL handles the request signing and pagination for you.

Absolutely. Where Langfuse provides native webhooks, you can use instant triggers to eliminate polling delay. For custom webhooks, WEXTL offers Custom Webhook node, so you can start automation whenever Langfuse or any other service sends event to webhook URL.

WEXTL is an AI-first visual canvas. The platform allows you to describe automation in plain text — and AI will wire nodes for Langfuse inside the workflow builder [4]. You can also use built-in workflow nodes (AI Agent, Human in the Loop, Condition, Aggregate Results) in pair with Langfuse modules to craft complex integration flows with branches, filtering, and conditional logic. There are absolutely no limits of what you can build.

References

List of sources used to gather information about the Langfuse integration.

Langfuse integration source code, module specs and credentials implementation.
Internal Langfuse integration documentation, structured input and output parameters, scopes, and authentication methods.
Official Langfuse vendor API documentation — used to verify field meanings, authentication flows, and anything our own specs don't fully capture.
WEXTL® Visual Workflow Builder, anonymized data from our technical support team.
WEXTL® Apps Library
WEXTL® Langfuse App Documentation
WEXTL® Workflow Scheduler
WEXTL® HTTP node documentation

Langfuse Integration

Supported triggers and actions for Langfuse integration