Schema | Celmis MCP Server

Celmis MCP Server

Official

Describes the environment variables required to run the server.

Name	Required	Description	Default
`CELMIS_MASTER_KEY`	No	Global admin password used with CELMIS_MASTER_EMAIL to sign in as admin.
`LITELLM_PROXY_URL`	No	When set together with LITELLM_MASTER_KEY, routes all model calls through a LiteLLM gateway.
`LITELLM_MASTER_KEY`	No	Master key for the LiteLLM gateway (set together with LITELLM_PROXY_URL).
`CELMIS_MASTER_EMAIL`	No	Global admin email used with CELMIS_MASTER_KEY to sign in as admin.
`CREDENTIAL_MASTER_KEY`	No	Key used to encrypt provider credentials before storing them in the database.
`CELMIS_DEPLOYMENT_MODE`	No	Whether deployment is single-tenant or multi-tenant (isolates workspaces from each other).	single_tenant
`REVIEW_TIMEOUT_SECONDS`	No	Wall clock for one review; past it the tail stages stand down and the comment says so.	900
`REVIEW_VERIFIER_ENABLED`	No	The LLM false-positive veto.	false
`CELMIS_JOB_LEASE_SECONDS`	No	Ceiling on worker silence before a job may be reclaimed.	600
`REVIEW_AGENT_CONCURRENCY`	No	Provider calls in flight per review.	3
`REVIEW_LLM_TIMEOUT_SECONDS`	No	One model call timeout. Raise to ~600 for a slow reasoning model.	300
`REVIEW_MAX_DIFF_SIZE_BYTES`	No	Larger diffs are refused, not truncated.	500000
`REVIEW_LLM_TIMEOUT_RETRY_FACTOR`	No	How much longer the retry gets after a timeout; 1.0 disables the widening.	2.0

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.

Server capabilities have not been inspected yet.

Functions exposed to the LLM to take actions

Name	Description
No tools

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

This server cannot be deployed

ActivityMaintained

ResponsivenessNo issues