Skip to main content
Glama
mirastacklabs-ai

Mirador MCP

Official

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
MCP_HTTP_HOSTNoCross-server alias for MIRADOR_SERVER__HOST.
MCP_HTTP_PATHNoCross-server alias for MIRADOR_SERVER__PATH.
MCP_HTTP_PORTNoCross-server alias for MIRADOR_SERVER__PORT.
MCP_TRANSPORTNoCross-server alias for MIRADOR_SERVER__TRANSPORT.
MCP_HTTP_LISTENNoAlternative cross-server alias for the listen address.
MCP_LISTEN_ADDRNoCross-server alias for MIRADOR_SERVER__LISTEN (host:port).
MIRADOR_LOGS__KINDNoLog backend kind: vlogs, loki, file, or none.
MIRADOR_DUCKDB_PATHNoPath to the DuckDB catalog file. Defaults to ~/.mirador/mirador.duckdb.
MIRADOR_FIXTURE_DIRNoPath to a committed fixture data directory to use instead of a live backend.
MIRADOR_SERVER__HOSTNoHost to bind for HTTP transports.
MIRADOR_SERVER__PATHNoHTTP path for the MCP endpoint. Defaults to /mcp.
MIRADOR_SERVER__PORTNoPort to bind for HTTP transports.
MIRADOR_TRACES__KINDNoTrace backend kind (e.g. vtraces) for VictoriaTraces.
MIRADOR_METRICS__KINDNoThe metrics backend kind (e.g. 'prometheus'). Required when connecting to a live Prometheus-compatible backend. Use together with MIRADOR_METRICS__BASE_URL.
MIRADOR_LOGS__BASE_URLNoOptional log backend URL.
MIRADOR_SERVER__LISTENNoHost:port to listen on for HTTP transports (e.g. 127.0.0.1:9001).
MIRADOR_TRACES__BASE_URLNoTrace backend URL (Jaeger-compatible API).
MIRADOR_METRICS__BASE_URLNoPrometheus-compatible query endpoint URL (e.g. http://localhost:8481/select/0/prometheus). Required when MIRADOR_METRICS__KIND is set.
MIRADOR_SERVER__STATELESSNoWhether to run stateless (true/false). Defaults to true.
MIRADOR_SERVER__TRANSPORTNoMCP transport: stdio, streamable-http, or sse. Defaults to stdio.
MIRADOR_TRACES__TLS_VERIFYNoWhether to verify TLS for the traces backend (true/false).
MIRADOR_SERVER__JSON_RESPONSENoWhether to use JSON responses (true/false). Defaults to true.
MIRADOR_BUDGET__MAX_POINTS_PER_RUNNoMaximum points allowed per run; raise for very large estates.

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