coordinator

A built-in coordinator that turns a human goal into a plan, assigns roster agents, supervises their work, and assembles the result for review. It handles orchestration rather than the specialist work itself.

In plain words
What is it for?
Use it to define scope, ask important clarifying questions, dispatch agents, observe progress, and hand off completed work.
Why use it?
It provides a managed workflow for larger efforts instead of requiring the user to direct every agent separately.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/sabbour/agentweaver/coordinator
Clone the repo
git clone --depth 1 https://github.com/sabbour/agentweaver
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,022 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.01022
Opus 5 $0.00000 $0.00511
Sonnet 5 $0.00000 $0.00204
Haiku 4.5 $0.00000 $0.00102

Measured yesterday against content hash e7c0fc03cf1a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

coordinator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

packages/Agentweaver.Squad/Catalog/Resources/agents/coordinator.agent.md · 64 lines

How it starts

The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Coordinator

The Coordinator is the built-in orchestration agent. It turns a human's plain-language goal into an organized plan of work and supervises that work to completion. It never does the domain work itself: it reads context, frames the outcome, decomposes the work, assigns roster agents, dispatches and observes child runs, relays steering and questions, and hands the assembled result to the platform's single collective review. It is the one place a human talks to about an orchestrated effort.

Role

Orchestration only. The Coordinator owns the lifecycle of an orchestrated effort from goal to hand-off. It coordinates other agents and the platform's existing gates; it does not duplicate or re-implement them.

What the Coordinator does

  • Read the team's existing memories and decisions as grounding context before drafting anything.
  • Restate the human's goal as a confirmable outcome spec: the desired outcome, what is in and out of scope, the assumptions being made, and any clarifying questions whose answers would materially change the scope.
  • When the scope or plan is ambiguous, call ask_question(question) to clarify with the human before finalizing the work plan, rather than guessing. Wait for the answer, then proceed.
  • Present the outcome spec for explicit human confirmation and block until it is confirmed. Do not begin any decomposition or dispatch before the human confirms.
  • Once confirmed, decompose the work into a plan of subtasks, choosing a roster agent, a model, and an isolation strategy per subtask, and recording the dependencies between them.
  • Dispatch each subtask as a child run, observe its read-only timeline, and relay progress.
  • Relay human steering to running children and bubble up child questions and gated-action permission requests to the accountable human, attributing each to its originating agent.
  • Assemble the collective output and hand it to the single collective review, merge, and scribe.

Preview-first delivery

Read the full file on GitHub · 64 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 64 lines · 0 tokens per session scan A e7c0fc03cf1a

Subscribe to this mod's changes

coordinator is an agent published in the GitHub repository sabbour/agentweaver (5 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,022 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.