Skip to main content
Glama
ph20
by ph20

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
HOSTNoApplication bind address (in Docker 0.0.0.0:8000).
PORTNoApplication bind port.
APP_ENVNodevelopment / production (fail-closed).
LOG_LEVELNoLogging level.
ACME_EMAILNoE-mail for Let's Encrypt.
LOG_FORMATNoJSON (default) or text; secrets are redacted.
MCP_DOMAINNoPublic hostname for Caddy and the Host check. Required in production (or use ALLOWED_HOSTS).
ALLOWED_HOSTSNoExtra Host values (comma-separated). Required in production if MCP_DOMAIN is not set.
AUTH_DISABLEDNotrue disables auth only in development; in production it is a startup error.
PROXY_HEADERSNoTrust X-Forwarded-* only from your own Caddy.
ALLOWED_ORIGINSNoExtra Origin values (comma-separated). CLI clients without an Origin pass; a foreign Origin is rejected (403), a foreign Host gets 421.
CACHE_MAX_BYTESNoMax cache bytes.
YOUTUBE_API_KEYNoGoogle key restricted to the YouTube Data API v3. Optional; without it search returns GOOGLE_API_NOT_CONFIGURED but transcripts keep working.
MCP_TOKEN_SHA256NoSHA-256 hex digest(s) of the bearer token, comma-separated for rotation. Required in production.
RATE_LIMIT_BURSTNoRate limit burst.
CACHE_MAX_ENTRIESNoMax cache entries.
MAX_RESPONSE_BYTESNoSize of one MCP result page (default 200,000).
FORWARDED_ALLOW_IPSNoTrust X-Forwarded-* only from your own Caddy (compose assigns the static address 172.28.0.10).
TOOL_TIMEOUT_SECONDSNoTime budget of one call.
TRANSCRIPT_MAX_BYTESNoMax transcript bytes.
UPSTREAM_MAX_RETRIESNoBounded retries with backoff/jitter.
RATE_LIMIT_PER_MINUTENoRate limit per minute.
MAX_REQUEST_BODY_BYTESNoMax HTTP request body bytes.
TRANSCRIPT_MAX_SEGMENTSNoMax transcript segments.
SEARCH_CACHE_TTL_SECONDSNoSearch cache TTL (default 300).
UPSTREAM_MAX_CONCURRENCYNoConcurrent upstream request limit.
GOOGLE_MAX_RESPONSE_BYTESNoMax Google API response bytes.
TRANSCRIPT_CACHE_TTL_SECONDSNoTranscript cache TTL (default 3600).
UPSTREAM_RETRY_BUDGET_SECONDSNoRetry budget in seconds.
UPSTREAM_QUEUE_TIMEOUT_SECONDSNoQueue timeout for upstream requests.

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