Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
AIGC_LITE_API_KEYNoOptional API key protecting all endpoints when enabled. If unset, the server runs in local single-tenant mode.
AIGC_LITE_LLM_MODELYesModel identifier to use for agent and chat requests.
AIGC_LITE_MASTER_KEYNoFernet master key used to encrypt credentials stored through the management interface. Required only if you plan to save model credentials or remote MCP header secrets in the database.
AIGC_LITE_LLM_API_KEYYesYour provider API key for the LLM.
AIGC_LITE_MCP_API_KEYNoCompatibility mode API key that maps to the default workspace for incoming MCP requests. Use user tokens or tenant API keys for multi-workspace deployments.
AIGC_LITE_LLM_BASE_URLYesBase URL of the OpenAI-compatible upstream API.
AIGC_LITE_TENANTS_JSONNoJSON array with tenant definitions, e.g. [{"id":"team-a","name":"Team A","api_key":"team-a-secret"}]. Enables multi-tenant API-key authentication and tenant-level data isolation.
AIGC_LITE_MAX_AGENT_STEPSNoMaximum number of agent model steps before a run is terminated with limit_reached.8
AIGC_LITE_MCP_SERVERS_JSONNoJSON array configuring remote MCP servers, e.g. [{"id":"research","url":"http://127.0.0.1:9000/mcp","workspace_id":"default","header_env":{"Authorization":"RESEARCH_MCP_AUTH"},"risk":"low"}].
AIGC_LITE_MAX_AGENT_TOOL_CALLSNoMaximum number of tool calls per agent run before termination.16
AIGC_LITE_MAX_AGENT_RUN_SECONDSNoMaximum wall-clock time in seconds for an agent run.300
AIGC_LITE_LEGACY_MCP_PROTOCOL_VERSIONNoProtocol version used only by the legacy /mcp-legacy debug endpoint. Does not affect the official /mcp Streamable HTTP endpoint.
AIGC_LITE_DEFAULT_TOOL_TIMEOUT_SECONDSNoDefault timeout in seconds for tool calls unless overridden by a tool decorator.30

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