Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
DEBUGNoEnable debug timing logs ('true' or '1').false
HTTP_PORTNoPort for HTTP transport.3000
QDRANT_URLNoQdrant server URL.http://localhost:6333
COHERE_API_KEYNoAPI key for Cohere.
OPENAI_API_KEYNoAPI key for OpenAI.
QDRANT_API_KEYNoAPI key for Qdrant authentication.
TRANSPORT_MODENoTransport mode: 'stdio' or 'http'.stdio
VOYAGE_API_KEYNoAPI key for Voyage AI.
CODE_BATCH_SIZENoNumber of chunks to embed in one batch.100
CODE_CHUNK_SIZENoMaximum chunk size in characters for code indexing.2500
CODE_ENABLE_ASTNoEnable AST-aware chunking using tree-sitter.true
EMBEDDING_MODELNoThe model name for embeddings. Provider-specific defaults apply if omitted.
DELETE_BATCH_SIZENoPaths per delete batch.500
CODE_CHUNK_OVERLAPNoOverlap between chunks in characters.300
CODE_CUSTOM_IGNORENoAdditional ignore patterns (comma-separated).
CODE_DEFAULT_LIMITNoDefault search result limit for code search.5
DELETE_CONCURRENCYNoParallel delete requests.8
EMBEDDING_BASE_URLNoCustom API URL for the embedding provider.
EMBEDDING_PROVIDERNoThe embedding provider to use: 'ollama', 'openai', 'cohere', or 'voyage'.ollama
MAX_IO_CONCURRENCYNoMaximum parallel file I/O operations during cache sync.50
EMBEDDING_DIMENSIONNoVector dimensions (auto-detected from model if possible).
PROMPTS_CONFIG_FILENoPath to prompts configuration JSON.prompts.json
EMBEDDING_BATCH_SIZENoNumber of texts per embedding request.64
EMBEDDING_CONCURRENCYNoParallel embedding requests (useful for multiple GPUs).1
EMBEDDING_RETRY_DELAYNoInitial retry delay in milliseconds.1000
QDRANT_BATCH_ORDERINGNoOrdering mode: 'weak', 'medium', or 'strong'.weak
CODE_CUSTOM_EXTENSIONSNoAdditional file extensions (comma-separated) for indexing.
HTTP_REQUEST_TIMEOUT_MSNoRequest timeout for HTTP transport in milliseconds.300000
CODE_ENABLE_GIT_METADATANoEnrich chunks with git blame information (author, dates, tasks).false
EMBEDDING_RETRY_ATTEMPTSNoNumber of retry attempts for embedding requests.3
QDRANT_FLUSH_INTERVAL_MSNoAuto-flush buffer interval for Qdrant (0 to disable).500
EMBEDDING_MAX_REQUESTS_PER_MINUTENoRate limit for embedding requests.

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/artk0de/TeaRAGs-MCP'

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