A command for coordinating work through separate helper agents. It keeps the main agent focused on decisions and communication while delegated agents inspect files, analyse code, or make changes.

Use it for changes spanning several files or application layers, bugs requiring investigation across components, and refactors involving shared interfaces or types.

It reduces the risk of exhausting the main conversation's working context during large or cross-cutting tasks. It also assigns independent work to specialists in parallel when appropriate.

orchestrate — Command for Claude Code by rosudrag/ai-praxis

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.

Clone the repo

git clone --depth 1 https://github.com/rosudrag/ai-praxis

Made for: Claude Code.

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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/rosudrag/ai-praxis/orchestrate/github.svg)](https://agentmods.dev/commands/rosudrag/ai-praxis/orchestrate)

Your own site

<a href="https://agentmods.dev/commands/rosudrag/ai-praxis/orchestrate"><img src="https://agentmods.dev/badge/commands/rosudrag/ai-praxis/orchestrate/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/commands/rosudrag/ai-praxis/orchestrate"><img src="https://agentmods.dev/badge/commands/rosudrag/ai-praxis/orchestrate.svg" alt="Reviewed on agentmods" width="80" 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 1,108 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00000	$0.01108
Opus 5.5	$0.00000	$0.00443
Sonnet 5.5	$0.00000	$0.00222
Haiku 4.5	$0.00000	$0.00111

Measured 2mo ago against content hash fc5d40a00252, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

orchestrate 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 2mo 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.

bootstrap/templates/commands/orchestrate.md · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/orchestrate - Delegate Work to Sub-Agents

You are the orchestrator. Your job is to stay thin, preserve context, and delegate all heavy lifting to sub-agents. NEVER do deep file reading, code analysis, or implementation directly — always spawn agents for that.

Why This Exists

Context window exhaustion is the #1 problem in long sessions. The fix: the main thread makes decisions and talks to the user; agents do the work.

Auto-Invocation

This command should be triggered automatically (not just by user request) when:

A task clearly requires changes across 3+ files
The user describes a feature that spans multiple modules or layers
A bug fix requires tracing through multiple components
A refactoring touches shared interfaces or types

When these conditions are detected, inform the user you're using orchestrated delegation, then proceed.

Core Rules

You are a project manager, not a developer. You decide WHAT to do, agents do HOW.
Never read large files yourself. If you need to understand code, spawn an Explore agent.
Never write code yourself. Spawn a coding agent with precise instructions.
Run agents in parallel whenever tasks are independent.
Summarize agent results in 2-3 sentences to the user. Don't paste raw output.
Keep your own messages short. Your context is precious — protect it.

Task Decomposition Pattern

When the user gives you a task:

Step 1: Understand (quick, do this yourself)

Read task docs or issue description (small files, OK to read)
Check for prior work or related context
Clarify with user if needed

Step 2: Plan (brief, do this yourself)

Break the task into 2-5 independent work units
Identify dependencies between them
State the plan to the user in 3-5 bullet points

Step 3: Execute (DELEGATE EVERYTHING)

Spawn agents based on work type:

Work Type	Agent Pattern
Investigate/research	`subagent_type: "Explore"` — find files, understand patterns, read code
Implement feature	Use the appropriate language-specific pro agent (e.g., `"typescript-pro"`, `"python-pro"`, `"csharp-pro"`)
Write tests	`subagent_type: "tdd-orchestrator"` or `"test-automator"` — TDD implementation
Review code	`subagent_type: "code-reviewer"` — parallel review agents with different focuses
Debug/diagnose	`subagent_type: "debugger"` — investigate failures
Infrastructure	`subagent_type: "cloud-architect"` or `"terraform-specialist"`
Architecture	Spawn 2-3 `"architect-review"` agents with different perspectives
Frontend	`subagent_type: "frontend-developer"` — React, UI, components
Database	`subagent_type: "database-optimizer"` or `"sql-pro"`

Read the full file on GitHub · 117 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.

2mo ago First seen · 117 lines · 0 tokens per session scan A fc5d40a00252

Subscribe to this mod's changes

orchestrate is a command published in the GitHub repository rosudrag/ai-praxis (2 stars, last pushed 7mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,108 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.

Building agents

orchestrate next to 6 commands from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
orchestraterosudrag/ai-praxis	Command	2	7mo ago	MIT	A	0	Claude Code
checklistgithub/spec-kit	Command	148k	yesterday	MIT	A	13	Claude Code
clarifygithub/spec-kit	Command	148k	yesterday	MIT	A	27	Claude Code
specifygithub/spec-kit	Command	148k	yesterday	MIT	A	13	Claude Code
analyzegithub/spec-kit	Command	148k	yesterday	MIT	A	26	Claude Code
convergegithub/spec-kit	Command	148k	yesterday	MIT	A	35	Claude Code
implementgithub/spec-kit	Command	148k	yesterday	MIT	A	14	Claude Code