Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
GROBID_URLNoURL of GROBID service for optional PDF paper metadata enrichment, e.g. http://127.0.0.1:8070
PDF_DOCLING_DEVICENoOverride Docling accelerator device, e.g. auto, cpu, or mps on Apple Silicon
GROBID_TIMEOUT_SECONDSNoTimeout for GROBID requests, default 20, recommended 120 for large papers
PDF_DOCLING_BATCH_SIZENoOverride Docling OCR/layout/table batch sizes, auto-derived by default
PDF_FORMULA_BATCH_SIZENoBatch size for formula processing, default auto (dynamically scaled)
PDF_DOCLING_NUM_THREADSNoOverride Docling CPU threads, auto-derived from CPU cores by default
MCP_EBOOK_INGEST_WORKERSNoNumber of parallel eager ingest workers for multiple new documents, default auto
PDF_PARSE_TIMEOUT_SECONDSNoTimeout for isolated Docling/Pix2Text PDF parse and visual extraction workers, default 1800
DOCLING_FORMULA_ENRICHMENTNoSet to true to enable Docling CodeFormulaV2 VLM enrichment (default false)
PDF_FORMULA_REQUIRE_ENGINENoSet to true to fail fast when Pix2Text is unavailable for formulas (default false)
MCP_EBOOK_CAPTURE_INCLUDE_QUERYNoSet to 1 to include raw query text in captures (default false)
MCP_EBOOK_CAPTURE_READING_SESSIONNoSet to 1 to capture read/search tool outputs for retrieval drift evaluation (default false)

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/ArtiPyHeart/mcp-ebook-read'

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