Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault

No arguments

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

Delegate coding work to Antigravity (agy) through the delegate tool. Every permission the agent requests is auto-approved, in every mode: agent/plan/ask reach any file the user account can, and mode instructs the agent rather than limiting it. Scope workspace tightly and review the git diff after every run — filesReportedByEditTools carries only what the agent's edit tools reported; the diff is authoritative.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
delegateA

Delegate a coding task to Antigravity's agy over ACP. Never shell out to agy — use this tool only. Auto-approves every permission the agent requests, in any mode and anywhere on disk. Clarifying questions arrive as prose in result — resume with resumeSessionId to answer. Leave model and effort unset unless the user explicitly asks. See the delegate skill.

cancelA

Best-effort cancel of an in-flight delegation by sessionId. session/cancel is advisory — the agent may finish the turn anyway, and the delegate result then carries cancelRequested. Status: cancelled, killed (agent process observed to exit), not-running (turn already ended, still resumable), not-found (id not in this process's recent session history).

doctorA

Setup diagnostics: plugin version, MCP client capabilities, agy launcher resolution. Run when delegation fails or agent.found is false.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

A4.1/5.0

Scored across 3 tools

Disambiguation5/5

Each tool maps to a clearly distinct action in the delegation lifecycle: start/resume a delegation (delegate), abort one (cancel), and diagnose setup failures (doctor). The descriptions explicitly scope when to use each, including the 'never shell out to agy' guardrail, so an agent can select without ambiguity.

Naming Consistency4/5

All names are single lowercase words with no mixed casing or prefixes, so the set is predictable. The only deviation is that 'doctor' is a noun while 'delegate' and 'cancel' are verbs, a minor break in the verb pattern.

Tool Count4/5

Three tools is minimal but well-matched to a narrow proxy purpose: one action tool, one lifecycle-control tool, one diagnostic. Nothing extraneous is present, though the surface is thin enough that a fourth helper (e.g. list sessions) could be justified.

Completeness4/5

The core lifecycle is covered: start/resume delegation, cancel an in-flight run, and troubleshoot setup. The main gap is discoverability — there is no way to list or query active/recent session IDs, so an agent must track sessionIds itself, and resume is only reachable by re-invoking delegate.

Maintenance

ActivityMaintained
ResponsivenessNo recent community issues