Skip to main content
Glama
Sync and MCP

Sync and MCP

by nejroseld

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
OBSI_HOSTNoListen address.127.0.0.1
OBSI_PORTNoListen port.8765
OBSI_CONFIGNoPath to a TOML configuration file; configuration is read from the [server] section.
OBSI_S3_SSENoServer-side encryption for S3 objects: AES256, aws:kms, or none; auto means AES256 for AWS and none for a custom endpoint.auto
OBSI_API_DOCSNoSet to 1 to enable /docs, /redoc, and /openapi.json for development.0
OBSI_DATA_DIRNoDirectory for the SQLite database and blobs../obsi-data
OBSI_FEATURESNoEnabled modules: ai, mcp, rag, all, or none. mcp and rag require ai.ai,mcp,rag
OBSI_S3_BUCKETNoBucket for attachments above the threshold; requires boto3 credentials.
OBSI_S3_PREFIXNoS3 object key prefix.obsi
OBSI_S3_REGIONNoS3 region.
OBSI_RAG_WORKERNoNumber of background embedding threads.1
OBSI_MAX_BODY_MBNoLimit for encrypted file uploads; the JSON API is limited to 1 MiB, login and registration to 16 KiB, MCP to 8 MiB.200
OBSI_S3_ENDPOINTNoS3 endpoint.
OBSI_CORS_ORIGINSNoExplicit comma-separated origin list; wildcard * is not accepted.app://obsidian.md,capacitor://localhost,http://localhost
OBSI_S3_KMS_KEY_IDNoKMS key ID for S3 server-side encryption when OBSI_S3_SSE is aws:kms.
OBSI_MAX_AI_FILE_MBNoLimit for one AI Available file.32
OBSI_VAULT_QUOTA_MBNoVault quota for ciphertext, AI Available files, and pending MCP edits; 0 means no quota.10240
OBSI_INVITE_TTL_DAYSNoInvitation lifetime.7
OBSI_S3_THRESHOLD_MBNoOffload threshold for files other than .md.10
OBSI_SESSION_TTL_DAYSNoSession lifetime after login.180
OBSI_SESSION_IDLE_DAYSNoIdle lifetime for sessions; at most 10 sessions live per user.30
OBSI_MAX_CONCURRENT_UPLOADSNoConcurrent uploads larger than 1 MiB; the rest wait in a queue.4
OBSI_EMBEDDING_ALLOW_PRIVATENoSet to 1 to allow http:// and private addresses for embeddings, such as local Ollama.0
OBSI_SYNC_REQUESTS_PER_MINUTENoRequest limit per authorized token per minute.6000

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