@jurislm/langfuse-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@@jurislm/langfuse-mcplist recent traces with high latency"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
@jurislm/langfuse-mcp
MCP (Model Context Protocol) server for Langfuse — provides 50 tools for prompt management, observability, instance/organization management, and metrics tracking via natural language.
Tools (50)
Prompt Management (6 tools)
listPrompts— List and filter prompts with pagination, labels, tagsgetPrompt— Fetch a specific prompt version with resolved dependenciescreateTextPrompt— Create a text prompt version (supports{{variable}}syntax)createChatPrompt— Create a chat prompt with role-based messagesgetPromptUnresolved— Fetch a prompt without resolving dependencies (for debugging composition)updatePromptLabels— Update labels for a prompt version (labels are unique across versions)
Traces (2 tools)
listTraces— List traces with pagination and optional filters (name, userId, tags, timestamps)getTrace— Get a single trace with full details, observations, and scores
Observations (2 tools)
listObservations— List observations (generations, spans, events) with filters by type and namegetObservation— Get a single observation with complete details (input, output, model, usage, duration)
Scores (2 tools)
createScore— Create a score for a trace or observation (for evals, feedback, quality metrics)listScores— List scores with pagination and optional filters
Sessions (1 tool)
listSessions— List sessions that group related traces together across time ranges
Datasets (8 tools)
listDatasets— List datasets for evals and benchmarking with search and paginationgetDataset— Get dataset details with metadata and item countscreateDataset— Create a new dataset for evals or benchmarkinglistDatasetItems— List items in a dataset with pagination (input/output pairs)createDatasetItem— Add an input/output pair to a dataset for eval testinggetDatasetItem— Get a single dataset item by IDlistDatasetRuns— List eval execution runs (traces using this dataset)getDatasetRun— Get details of a dataset run (eval execution)
Metrics (2 tools)
getDailyMetrics— Get daily metrics for traces and observations (latency, tokens, scores, cost)getUsageSummary— Get usage summary for a time period (total traces, observations, tokens, cost)
Score Configurations (4 tools)
listScoreConfigs— List all score configurations (custom score names and data types)getScoreConfig— Get details of a score configuration including name, data type, and categoriescreateScoreConfig— Create a new score configuration for custom scoring metricsupdateScoreConfig— Update an existing score configuration
Project (1 tool)
getProject— Get current project details (name, API keys, limits, usage)
Instance Management (8 tools)
listInstances— List all Langfuse instances with paginationgetInstance— Get details of a specific instancecreateInstance— Create a new instanceupdateInstance— Update an existing instancedeleteInstance— Delete an instancelistInstanceEvents— List events for an instancegetInstanceStatus— Get health and status informationconfigureInstanceSettings— Configure advanced settings
Organization Projects (5 tools)
listOrganizationProjects— List all projects within an organizationgetOrganizationProject— Get details of a specific projectcreateOrganizationProject— Create a new project in an organizationupdateOrganizationProject— Update an existing projectdeleteOrganizationProject— Delete a project from an organization
Organization API Keys (3 tools)
listOrganizationApiKeys— List all API keys for an organizationcreateOrganizationApiKey— Create a new API keydeleteOrganizationApiKey— Delete an API key
Organization Memberships (6 tools)
listOrganizationMembers— List all members of an organizationaddOrganizationMember— Add a new member to an organizationupdateOrganizationMember— Update a member's role or permissionsremoveOrganizationMember— Remove a member from an organizationlistProjectMembers— List members with access to a specific projectupdateProjectMember— Update a member's access level for a project
Related MCP server: langfuse-mcp
Setup
Environment Variables
# Public API (Required for Prompt Management, Traces, Observations, Scores, Sessions, etc.)
LANGFUSE_PUBLIC_KEY=pk-lf-...
LANGFUSE_SECRET_KEY=sk-lf-...
# Instance/Organization Management APIs (Required for Instance and Organization tools)
LANGFUSE_ADMIN_API_KEY=adm-lf-...
# Optional: Organization-specific API key for additional org-level tools
LANGFUSE_ORG_API_KEY=org-lf-...
# Optional: API Server (defaults to https://cloud.langfuse.com)
LANGFUSE_HOST=https://cloud.langfuse.com
# US region: https://us.cloud.langfuse.comUsage with Claude Code (via npx)
Add to your MCP configuration (.mcp.json or ~/.claude/settings.json):
{
"mcpServers": {
"langfuse": {
"command": "npx",
"args": ["-y", "@jurislm/langfuse-mcp@latest"],
"env": {
"LANGFUSE_PUBLIC_KEY": "pk-lf-...",
"LANGFUSE_SECRET_KEY": "sk-lf-...",
"LANGFUSE_HOST": "https://cloud.langfuse.com"
}
}
}
}Note: Runtime requires only Node.js >= 18 — the published binary runs under plain
node, sonpxworks without any extra tooling. Bun is needed only for local development (see Development), never to run this server.
Usage with Claude Code Plugin (jurislm-tools)
If you use the jurislm-tools Claude Code plugin, jt:langfuse is included:
/plugin marketplace update jurislm-toolsThen set environment variables in ~/.zshenv:
export LANGFUSE_PUBLIC_KEY=pk-lf-...
export LANGFUSE_SECRET_KEY=sk-lf-...
export LANGFUSE_HOST=https://us.cloud.langfuse.comDevelopment
bun install
bun run dev # Run locally (stdio transport)
bun run build # Build to dist/
bun run typecheck # TypeScript check
bun run lint # ESLint (max-warnings=0)License
MIT
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/jurislm/langfuse-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server