# Langfuse integration

> Open-source LLM observability and engineering platform: traces, observations, scores, prompts, datasets, experiments, evaluators and models via the Langfuse public API.

- Authentication: apiKey

## Tools (84)

- **Add Dashboard Placement (Unstable)**: Add a widget or preset placement to a dashboard grid
- **Create Annotation Queue**: Create an annotation queue
- **Create Annotation Queue Assignment**: Create an assignment for a user to an annotation queue
- **Create Annotation Queue Item**: Add an item to an annotation queue
- **Create Comment**: Create a comment on a trace, observation, session or prompt
- **Create Dashboard (Unstable)**: Create a dashboard
- **Create Dashboard Widget (Unstable)**: Create a standalone dashboard widget (chart definition)
- **Create Dataset**: Create a dataset
- **Create Dataset Item**: Create a dataset item
- **Create Evaluation Rule**: Create an evaluation rule using stable evaluator identifiers
- **Create Evaluator**: Create an evaluator (LLM-as-a-judge or code) in the authenticated project
- **Create Model**: Create a model definition with pricing
- **Create Prompt**: Create a new version for the prompt with the given name (text or chat)
- **Create Score**: Create one score or a batch of scores asynchronously
- **Create Score Config**: Create a score configuration that defines the structure of scores
- **Create Skill Version (Unstable)**: Create a skill version
- **Delete Annotation Queue Assignment**: Delete an assignment for a user to an annotation queue
- **Delete Annotation Queue Item**: Remove an item from an annotation queue
- **Delete Dashboard Placement (Unstable)**: Remove a placement from a dashboard grid without deleting the widget
- **Delete Dashboard (Unstable)**: Delete a dashboard
- **Delete Dashboard Widget (Unstable)**: Delete a dashboard widget
- **Delete Dataset Item**: Delete a dataset item and all its run items
- **Delete Evaluation Rule**: Delete an evaluation rule
- **Delete Evaluator**: Delete an evaluator and all of its stored versions
- **Delete LLM Connection**: Delete an LLM connection by id
- **Delete Model**: Delete a model
- **Delete Multiple Traces**: Delete multiple traces
- **Delete Prompt**: Delete prompt versions (all versions if no version or label is given)
- **Delete Score**: Delete a score (supports both trace and session scores)
- **Delete Skill Version (Unstable)**: Delete one immutable skill version
- **Delete Trace**: Delete a specific trace
- **Export OpenTelemetry Traces**: OTLP/HTTP trace ingestion endpoint (real-time write path)
- **Get Annotation Queue**: Get an annotation queue by ID
- **Get Annotation Queue Item**: Get a specific item from an annotation queue
- **Get Comment**: Get a comment by id
- **Get Current Project**: Get the project associated with the project-scoped API key
- **Get Dashboard (Unstable)**: Get a dashboard by id
- **Get Dashboard Widget (Unstable)**: Get a dashboard widget by id
- **Get Dataset**: Get a dataset
- **Get Dataset Item**: Get a dataset item
- **Get Evaluation Rule**: Get one evaluation rule by its stable identifier
- **Get Evaluator**: Get one evaluator by its stable identifier
- **Get Media**: Get a media record
- **Get Media Upload URL**: Get a presigned upload URL for a media record
- **Get Model**: Get a model
- **Get Prompt**: Get a prompt by name, version or label
- **Get Score Config**: Get a score config
- **Get Skill File Contents (Unstable)**: Read a batch of skill file contents by SHA-256 hash
- **Get Skill (Unstable)**: Resolve a skill's metadata and file manifest by version or label
- **Health Check**: Check health of API and database
- **List Annotation Queue Items**: Get items for a specific annotation queue
- **List Annotation Queues**: Get all annotation queues
- **List Comments**: Get all comments
- **List Dashboards (Unstable)**: List dashboards, most recently updated first
- **List Dashboard Widgets (Unstable)**: List dashboard widgets, most recently updated first
- **List Dataset Items**: Get dataset items, optionally as of a point in time
- **List Datasets**: Get all datasets
- **List Evaluation Rules**: List evaluation rules in newest-first creation order
- **List Evaluators**: List evaluators in newest-first creation order
- **List Evaluator Versions**: List an evaluator's version history in newest-first order
- **List Experiment Items**: List experiment items with cursor-based pagination
- **List Experiments**: List experiments with cursor-based pagination
- **List LLM Connections**: Get all LLM connections in a project
- **List Models**: Get all models
- **List Observations (v2)**: List observations with cursor-based pagination and field selection (real-time)
- **List Prompts**: Get a list of prompt names with versions and labels
- **List Score Configs**: Get all score configs
- **List Scores (v3)**: Get a list of scores with a polymorphic value field
- **List Skills (Unstable)**: List skills with latest-version metadata, shared tags and production version
- **Query Metrics (v2)**: Get aggregate metrics from the project using a JSON query object (real-time)
- **Set Skill Labels (Unstable)**: Replace the labels on a skill version, moving them from other versions
- **Submit Feedback**: Submit user-approved feedback about Langfuse skills, MCP tools, CLI, docs or public API
- **Update Annotation Queue Item**: Update an annotation queue item
- **Update Dashboard Placement (Unstable)**: Move or resize a dashboard placement
- **Update Dashboard (Unstable)**: Update a dashboard's name, description, definition, or filters
- **Update Dashboard Widget (Unstable)**: Update a dashboard widget
- **Update Evaluation Rule**: Update an evaluation rule by its stable identifier
- **Update Evaluator**: Update an evaluator's metadata or replace its definition
- **Update Media Upload Status**: Patch a media record with the result of its upload
- **Update Prompt Version Labels**: Update labels for a specific prompt version
- **Update Score Config**: Update a score config
- **Update Skill (Unstable)**: Replace the shared tags across all versions of a skill
- **Upsert LLM Connection**: Create or update an LLM connection (upserted on provider)
- **Upsert Model**: Create or replace a project-owned model using its id

Connect Langfuse in General Input: https://www.generalinput.com/apps/langfuse