Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
DATABASE_URLYesPostgreSQL connection string for the super_mcp database. Example: postgresql://postgres@localhost:5432/super_mcp
SUPER_MCP_FULLNoSet to '1' to ingest more stores within the configured region (no per-chain count cap). Default is off.0
SUPER_MCP_NO_CAPNoSet to '1' to ingest all in-region stores (implies all Cerberus chains). Default is off.0
SUPER_MCP_CONCURRENCYNoConcurrency for price file ingestion. Default is 12, maximum 48.12
SUPER_MCP_EMBED_MODELNoActive embedding generation model. Default is multilingual MiniLM.multilingual MiniLM
SUPER_MCP_EMBED_BACKENDNoEmbedding backend; set to 'hasher' for deterministic fallback embedder (useful in tests/CI).
SUPER_MCP_REGION_FILTERNoSet to '0' to disable the region filter (Gush Dan–Sharon, Jerusalem, Haifa, Beersheva). Default is on.1
MAX_ANONYMOUS_IP_BUCKETSNoMaximum number of IP buckets tracked for anonymous rate limiting. Behind this cap, unseen addresses are limited by the global ceiling only.
SUPER_MCP_ALLOW_ANONYMOUSNoSet to '1' to allow keyless (anonymous) access to the shopping surface. Default is off.0
SUPER_MCP_SEMANTIC_BASKETNoMaster kill switch for semantic basket features. Set to '0' to disable recall, policy, and V2 shadow. Default is on.1
SUPER_MCP_SEMANTIC_SHADOWNoSet to '1' to log lexical vs semantic pick disagreements. Default is off.0
BASKET_CONTINUATION_SECRETYesSecret used for signed basket continuations. Must be at least 32 bytes.
SUPER_MCP_ONTOLOGY_VERSIONNoActive ontology id. Default is 'he-retail-v1'.he-retail-v1
SUPER_MCP_SEMANTIC_V2_POLICYNoSet to '0' to disable data-driven constraint gating. Default is on when basket is on.1
SUPER_MCP_SEMANTIC_V2_RECALLNoSet to '0' to disable query-vector recall / RRF. Default is on when basket is on.1
SUPER_MCP_SEMANTIC_V2_SHADOWNoSet to '1' to compute V2 recall+policy for logs but return lexical / pre-policy results. Default is off.0
SUPER_MCP_DETERMINISTIC_FIRSTNoSet to '0' to disable deterministic-first cascade (legacy blended ranking). Default is '1' when semantic basket is on.1
SUPER_MCP_SKIP_SEMANTIC_DRAINNoSet to '1' to skip post-ingest drain of semantic index dirty queue.0
SUPER_MCP_ANONYMOUS_RATE_LIMITNoPer-address rate limit for anonymous clients, per minute. Default is 30.30
SUPER_MCP_ALLOW_MCP_QUERY_API_KEYNoSet to '1' to allow API keys via query string for legacy MCP clients. Never enable on public hosts. Default is off.0
SUPER_MCP_ANONYMOUS_GLOBAL_RATE_LIMITNoGlobal rate limit for all keyless traffic, per minute. Default is 600.600

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

Latest Blog Posts

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/nitaiaharoni1/super-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server