Langfuse MCP server
{
"mcpServers": {
"pulsemcp-langfuse": {
"command": "npx",
"args": [
"-y",
"langfuse-observability-mcp-server@latest"
],
"env": {
"AGENT_ORCHESTRATOR_BASE_URL": "https://...",
"FLY_IO_API_TOKEN": "<fly_io_api_token>"
}
}
}
}{
"mcpServers": {
"pulsemcp-langfuse": {
"command": "npx",
"args": [
"-y",
"langfuse-observability-mcp-server@latest"
],
"env": {
"AGENT_ORCHESTRATOR_BASE_URL": "https://...",
"FLY_IO_API_TOKEN": "<fly_io_api_token>"
}
}
}
}{
"mcpServers": {
"pulsemcp-langfuse": {
"command": "npx",
"args": [
"-y",
"langfuse-observability-mcp-server@latest"
],
"env": {
"AGENT_ORCHESTRATOR_BASE_URL": "https://...",
"FLY_IO_API_TOKEN": "<fly_io_api_token>"
}
}
}
}{
"servers": {
"pulsemcp-langfuse": {
"command": "npx",
"args": [
"-y",
"langfuse-observability-mcp-server@latest"
],
"env": {
"AGENT_ORCHESTRATOR_BASE_URL": "${input:agent_orchestrator_base_url}",
"FLY_IO_API_TOKEN": "${input:fly_io_api_token}"
}
}
},
"inputs": [
{
"id": "agent_orchestrator_base_url",
"type": "promptString",
"description": "Agent orchestrator base url",
"password": true
},
{
"id": "fly_io_api_token",
"type": "promptString",
"description": "Fly io api token",
"password": true
}
]
}{
"mcpServers": {
"pulsemcp-langfuse": {
"command": "npx",
"args": [
"-y",
"langfuse-observability-mcp-server@latest"
],
"env": {
"AGENT_ORCHESTRATOR_BASE_URL": "https://...",
"FLY_IO_API_TOKEN": "<fly_io_api_token>"
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y langfuse-observability-mcp-serverversion 0.1.3
What it does
This MCP server provides access to Langfuse, an LLM observability platform for tracing and analyzing LLM observations. It is for developers who use Langfuse to monitor and debug their LLM applications.
- Query and retrieve LLM traces from Langfuse to analyze model behavior and performance.
- Access observation data to debug issues in LLM application workflows.
- Integrate Langfuse trace analysis into AI agent workflows for real-time observability.
This entry comes from the registry but points to no public repository, so there is nothing to check: no commit history, no source, no tool list. It is listed for completeness and left out of search engines.
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.