Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
LANGFUSE_HOSTNoLangfuse instance URL. For self-hosted instances use your own instance URL. Defaults to https://cloud.langfuse.com.https://cloud.langfuse.com
LANGFUSE_PUBLIC_KEYNoLangfuse public API key (e.g. pk-...). Precedence: CLI flags (--public-key) > environment variables > the selected profile > defaults.
LANGFUSE_SECRET_KEYNoLangfuse secret API key (e.g. sk-...). Precedence: CLI flags (--secret-key) > environment variables > the selected profile > defaults.
LANGFUSE_MCP_PROFILENoProfile to load from the profiles file; same as the --profile flag. Selects a Langfuse project's credentials from the profiles file.
LANGFUSE_MAX_AGE_DAYSNoCaps the lookback window for time-based tools (fetch_traces, fetch_observations, etc.). Set to match your Langfuse instance's data retention, e.g. 30 if retention is 30 days.7
LANGFUSE_MCP_READ_ONLYNoSet to true to disable all write operations for safer read-only access (disables create/update/delete tools for prompts, datasets, and annotation queues). Equivalent to the --read-only flag.false
LANGFUSE_MCP_PROFILES_FILENoPath of the TOML profiles file; same as the --profiles-file flag. Defaults to $XDG_CONFIG_HOME/langfuse-mcp/profiles.toml (or ~/.config/langfuse-mcp/profiles.toml when XDG_CONFIG_HOME is not set).$XDG_CONFIG_HOME/langfuse-mcp/profiles.toml
LANGFUSE_MCP_DEFAULT_OUTPUT_MODENoSets the MCP-exposed default output_mode so clients that omit the parameter automatically use your preferred mode. Supported values: compact, full_json_string, full_json_file. Equivalent to the --default-output-mode flag. Clients can still override it per call by passing output_mode explicitly.
LANGFUSE_MCP_TRACE_TIMEOUT_SECONDSNoPer-request read timeout in seconds for single-trace fetches (fetch_trace). Raise it if large traces with include_observations=True time out. Must be a positive integer.120

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Server capabilities have not been inspected yet.

Tools

Functions exposed to the LLM to take actions

NameDescription

No tools

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources