Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
ATTEND_HOSTNoHost the attend serve listener binds to. Replaced by ATTEND_SOCKET when a socket path is given.127.0.0.1
ATTEND_PORTNoPort the attend serve listener binds to. Port 0 picks a free one. Replaced by ATTEND_SOCKET when a socket path is given.8787
ATTEND_JUDGENoWhich judge to use. `rule` interrupts on a flush and queues otherwise, with no network. `jev` reads TYPESAFE_API_KEY. `decisions` reads OPENAI_API_KEY and needs ATTEND_JUDGE_MODEL.rule
ATTEND_KINDSNoEvent name to kind, as inline JSON or a file, e.g. `{"user.said":"flush","build.failed":"closing"}`. Kinds map to settle roles: closing, flush, continuous, passive, consequence or context.
ATTEND_STORENoWhere attend keeps state. One of `memory`, `sqlite:<file>`, `postgres:<url>` or `redis:<url>`. Postgres and Redis need the `pg` or `redis` package installed.memory
ATTEND_TOKENNoRequire `Authorization: Bearer <token>` on ingest and MCP.
ATTEND_SOCKETNoUnix socket path for the attend serve listener; replaces host and port.
OPENAI_API_KEYNoYour OpenAI API key, read by the `decisions` judge.
TYPESAFE_API_KEYNoYour TypeSafe key, read by the `jev` judge.
ATTEND_JUDGE_MODELNoModel id required by the `decisions` judge (the `--judge-model` flag).
ATTEND_DEFAULT_KINDNoKind for event names that ATTEND_KINDS does not list.consequence
ATTEND_ACT_THRESHOLDNoAct threshold in (0, 1]. The route becomes an acting route (interrupt or ask, whichever is likelier and offered) exactly when p(interrupt) + p(ask) reaches the threshold, and a quieter route otherwise. Off by default, and only applies to judgments that carry real probabilities.

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