Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
S2_API_KEYNoOptional Semantic Scholar API key for higher rate limits
MAX_RETRIESNoMaximum number of retries for HTTP requests3
GITHUB_TOKENNoOptional GitHub token for higher API rate limits
HTTP_TIMEOUTNoTimeout in seconds for HTTP requests60
OLLAMA_MODELNoOllama model namemistral
ARXIV_DB_PATHNoPath to the SQLite workflow database${ARXIV_DOWNLOAD_DIR}/arxiv_mcp.db
DIFF_CACHE_TTLNoTTL in seconds for diff cache (default 24 hours)86400
LINK_CACHE_TTLNoTTL in seconds for code link cache (default 48 hours)172800
ARXIV_CACHE_DIRNoDirectory for caching downloaded PDFs./downloads
ARXIV_CACHE_TTLNoTTL in seconds for cache entries (default 7 days)604800
ARXIV_KEEP_PDFSNoWhether to keep PDFs after parsingfalse
EMBEDDING_MODELNoEmbedding model for semantic similaritysentence-transformers/all-MiniLM-L6-v2
OLLAMA_BASE_URLNoBase URL for Ollama serverhttp://localhost:11434
REPRO_CACHE_TTLNoTTL in seconds for reproducibility cache (default 7 days)604800
ARXIV_RATE_LIMITNoRate limit for arXiv API requests per second1
GITHUB_MAX_FILESNoMaximum number of files to analyze from a GitHub repository20
CHUNK_SIZE_TOKENSNoToken chunk size for context building800
ARXIV_DOWNLOAD_DIRNoDirectory for downloaded PDFs (used in MCP config)${workspaceFolder}/downloads
CITATION_CACHE_TTLNoTTL in seconds for citation cache (default 24 hours)86400
SEMANTIC_INDEX_DIRNoDirectory for semantic index storage${ARXIV_DOWNLOAD_DIR}/semantic_index
CHUNK_OVERLAP_TOKENSNoToken overlap between chunks100
ARXIV_RATE_LIMIT_DELAYNoDelay in seconds between arXiv API calls3.0
CONTRIBUTION_CACHE_TTLNoTTL in seconds for contribution cache (default 7 days)604800
GITHUB_MAX_FILE_SIZE_KBNoMaximum file size in KB for GitHub file analysis50

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/Aldrin-Joan/paperstack'

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