orchestrate

A workflow coordinator that plans a software task and assigns implementation work to different agent models. It uses higher-tier models for planning and harder work, and another model for routine implementation.

In plain words
What is it for?
Plan, decompose, coordinate, and integrate software builds or changes from a task description or plan file.
Why use it?
It separates project planning from hands-on coding so larger changes can be broken into manageable pieces and delegated.

Skill for Claude CodeCodex

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 skills/drn/dots/orchestrate
Any agent
npx skills add drn/dots --skill orchestrate
Clone the repo
git clone --depth 1 https://github.com/drn/dots

Made for: Claude Code, Codex.

Per session 88 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,711 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 $0.00088 $0.02711
Opus 5 $0.00044 $0.01355
Sonnet 5 $0.00018 $0.00542
Haiku 4.5 $0.00009 $0.00271

Measured 2d ago against content hash cdfdb31abd0d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, 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 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.

agents/skills/orchestrate/SKILL.md · 164 lines

How it starts

The opening of the file, as written. The whole thing — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Orchestrate: Tiered Planning + Implementation

Run the expensive top-tier session model only where it earns its cost — planning, decomposition, orchestration, and final integration — and fan implementation out through the Workflow tool to model-tiered subagents: Sonnet for routine work, Opus for complex work.

The top-tier planner/orchestrator is the session model — Fable when it is available, Opus when Fable is not. Planning runs in the main loop, so it automatically uses whichever model the session is on; no detection step is needed.

Invoking this skill is the explicit opt-in the Workflow tool requires.

Arguments

  • $ARGUMENTS — required: description of the feature or change to build, or a path to a plan/spec file.

If no arguments are provided, ask the user what to build and stop.

Context

  • Current branch: !git branch --show-current 2>/dev/null | head -1
  • Git status: !git status --short 2>/dev/null | head -20
  • Project type: !find . -maxdepth 1 \( -name go.mod -o -name package.json -o -name Cargo.toml -o -name pyproject.toml -o -name Gemfile -o -name Makefile \) 2>/dev/null | head -6
  • Recent commits: !git log --oneline -5 2>/dev/null | head -5

Step 1: Plan in the Main Loop

You (the session model) are the planner and orchestrator. Do NOT delegate planning to a cheaper model, and do NOT implement the tasks yourself.

  1. Understand the task. Explore the codebase and read the files the change will touch. For large or unfamiliar codebases, fan out scouting to Explore subagents first.
  2. Decompose into work items. Each work item needs:
    • id — short slug, e.g. api-endpoint
    • prompt — fully self-contained: file paths, existing conventions to follow, exact success criteria, and the instruction to run a sanity check (compile/lint) before finishing. Subagents have no conversation history; the prompt is everything they know.
    • tiersonnet or opus (rubric below)
    • files — the files this item owns

Read the full file on GitHub · 164 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. 2d ago First seen · 164 lines · 88 tokens per session scan A cdfdb31abd0d

Subscribe to this mod's changes

orchestrate is a skill published in the GitHub repository drn/dots (23 stars, last pushed 4d ago), licensed MIT. It adds 88 tokens to every session and 2,711 once invoked, about $0.0004 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 2026-08-30.