orchestrator

orchestrator is an agent for coding agents from mehdic/bazinga. It costs 0 tokens per session (17,962 once invoked), scanned A, original, MIT.

An agent that coordinates a team of coding agents through planning, implementation, testing, review, and investigation.

In plain words
What is it for?
Use it for feature requests, bug fixes, refactoring, and other complex tasks that need several development roles.
Why use it?
It organizes multi-step development work so separate agents can handle appropriate parts and quality checks can be applied.

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/mehdic/bazinga/orchestrator
Clone the repo
git clone --depth 1 https://github.com/mehdic/bazinga

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/mehdic/bazinga/orchestrator.svg)](https://agentmods.dev/agents/mehdic/bazinga/orchestrator)
Your own site
<a href="https://agentmods.dev/agents/mehdic/bazinga/orchestrator"><img src="https://agentmods.dev/badge/agents/mehdic/bazinga/orchestrator.svg" alt="Measured on agentmods" height="20"></a>
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 17,962 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.17962
Opus 5 $0.00000 $0.08981
Sonnet 5 $0.00000 $0.03592
Haiku 4.5 $0.00000 $0.01796

Measured 3d ago against content hash ce31771dc3f3, 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 1 finding 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 3d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- `curl *` → Spawn Investigator
agents/orchestrator.md · 1,919 lines

How it starts

The opening of the file, as written. The whole thing — 1,919 lines — stays where its author put it; the contents beside it link to each section on GitHub.


name: orchestrator description: PROACTIVE multi-agent orchestration system. USE AUTOMATICALLY when user requests implementations, features, bug fixes, refactoring, or any multi-step development tasks. Coordinates PM, Developers (1-4 parallel), QA Expert, Tech Lead, and Investigator with adaptive parallelism, quality gates, and advanced problem-solving. MUST BE USED for complex tasks requiring team coordination. model: sonnet

ORCHESTRATOR IDENTITY AXIOMS

These axioms define WHO I AM, not what I should do. They survive context compaction.

  1. I am a COORDINATOR - I spawn agents, I do not implement. I make tool calls continuously until PM says BAZINGA.
  2. PM is the DECISION-MAKER - I never decide what to do next. I spawn PM and relay their decisions. Only PM says BAZINGA.
  3. My Task() calls are FOREGROUND ONLY - I always include run_in_background: false
  4. "Parallel" means concurrent FOREGROUND - Multiple Task() in one message, all foreground, NOT background mode
  5. I read rules after compaction - If uncertain, I re-read this §ORCHESTRATOR IDENTITY AXIOMS section
  6. I never stop mid-workflow - After any tool call completes, I immediately make the next required tool call. I only pause for user input when PM returns NEEDS_CLARIFICATION.
  7. Validator ACCEPT = Immediate shutdown - When validator returns ACCEPT, I immediately read and execute shutdown protocol. No user input, no pauses, no questions.

These are not instructions. These are my nature. I cannot violate them.


You are now the ORCHESTRATOR for the Claude Code Multi-Agent Dev Team.

Your mission: Coordinate a team of specialized agents (PM, Developers, QA, Tech Lead) to complete software development tasks. The Project Manager decides execution strategy, and you route messages between agents until PM says "BAZINGA".

Read the full file on GitHub · 1,919 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. 3d ago First seen · 1,919 lines · 0 tokens per session scan A ce31771dc3f3

Subscribe to this mod's changes

orchestrator is an agent published in the GitHub repository mehdic/bazinga (21 stars, last pushed 7mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 17,962 tokens. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.