orchestrator

A set of rules for an orchestrator that coordinates coding agents without writing the product code itself. It assigns work through tasks and passes handoffs through shared work items.

In plain words
What is it for?
Use it to manage project stages, delegate development and review, track requests, and require approval before advancing.
Why use it?
It keeps coordination separate from implementation and makes transfers between agents traceable.

Cursor rule for Cursor

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 rules/naimax/agent-harness-reference/orchestrator
Clone the repo
git clone --depth 1 https://github.com/naimax/agent-harness-reference

Made for: Cursor.

Per session 270 This file is loaded in full into every session.
When invoked 270 The same file — it is already loaded in full.
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 $0.00270 $0.00270
Opus 5 $0.00135 $0.00135
Sonnet 5 $0.00054 $0.00054
Haiku 4.5 $0.00027 $0.00027

Measured 2d ago against content hash fd1561f318f2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

orchestrator 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 2d ago.

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.

.cursor/rules/orchestrator.mdc · 29 lines

What it actually says

Оркестратор

  1. Ты — оркестратор. Не пишешь project/src/** и project/docs/**. Только Task(...).
  2. Веди запрос: project/status/req-*.mdcurrent_stage.
  3. Межфабричный handoff — только через control-plane/work-items/ + validate_wi.py.

Стадии (сокращённо)

Исполнитель Ревьюер Артефакт
developer code-reviewer project/src/**
verifier отчёт + свежий прогон

Следующая стадия — только после APPROVED (sdlc-gate.mdc).

Work Item

Handoff между конвейерами = WI в control-plane/. Обновление markdown без WI — не передача работы.

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. 2d ago First seen · 29 lines · 270 tokens per session scan A fd1561f318f2

Subscribe to this mod's changes

orchestrator is a cursor rule published in the GitHub repository naimax/agent-harness-reference (8 stars, last pushed 1mo ago), licensed MIT. It adds 270 tokens to every session, about $0.0014 per session on Opus 5. 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.