setup-worker

setup-worker is a command for coding agents from divyamrastogi/model-orchestration. It costs 18 tokens per session (514 once invoked), scanned A, original, MIT.

A command for configuring model workers, which are separate AI processes used by an orchestrator to perform tasks. It supports preset providers and any provider with an Anthropic-compatible API.

In plain words
What is it for?
Use it to create one or more workers from known provider presets or configure a custom compatible provider, including local Ollama models.
Why use it?
It removes the need to manually assemble worker names, provider settings, endpoints, and model mappings.

Command

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the model-orchestration plugin — 1 skill, 2 commands, 1 agent, 1 hook shipped together

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 commands/divyamrastogi/model-orchestration/setup-worker
Clone the repo
git clone --depth 1 https://github.com/divyamrastogi/model-orchestration

Or install model-orchestration, the plugin that ships this one along with the rest of its 1 skill, 2 commands, 1 agent, 1 hook.

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 setup-worker

README.md
[![agentmods](https://agentmods.dev/badge/commands/divyamrastogi/model-orchestration/setup-worker.svg)](https://agentmods.dev/commands/divyamrastogi/model-orchestration/setup-worker)
Your own site
<a href="https://agentmods.dev/commands/divyamrastogi/model-orchestration/setup-worker"><img src="https://agentmods.dev/badge/commands/divyamrastogi/model-orchestration/setup-worker.svg" alt="Measured on agentmods" height="20"></a>
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 514 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 $0.00018 $0.00514
Opus 5 $0.00009 $0.00257
Sonnet 5 $0.00004 $0.00103
Haiku 4.5 $0.00002 $0.00051

Measured 3d ago against content hash 78fe4715823a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

setup-worker 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.

**Ollama preset:** first check the server is up (`curl -s localhost:11434/api/version`)
commands/setup-worker.md · 36 lines

What it actually says

Set up one or more model workers using the orchestrating-model-workers skill (read its SKILL.md in this plugin before starting).

Arguments given: "$ARGUMENTS"

  1. Run "${CLAUDE_PLUGIN_ROOT}/scripts/setup-worker.sh" presets to load the known providers.
  2. If no provider was specified in the arguments: ask the user which provider(s) to configure using the AskUserQuestion tool — one option per preset (label + note as description, multiSelect: true) — rather than making them type URLs. If they pick "Other", ask for the provider name and look up its Anthropic-compatible endpoint and model names in its docs (web search if needed).
  3. For each chosen provider, run: "${CLAUDE_PLUGIN_ROOT}/scripts/setup-worker.sh" create <name> --provider <id> Default the worker name to the preset id unless the user gave one. Ollama preset: first check the server is up (curl -s localhost:11434/api/version) and run ollama list, then ask the user which local model to map the tiers to (offer instruct/tool-capable models; warn against -mlx variants — see the skill's local-models caveats). Pass it via --sonnet <model> and set the key yourself: printf 'ollama' | setup-worker.sh set-key <name>.
  4. Show the user where to get the key (the preset's key_url) and exactly how to add it (the two options the script printed). NEVER ask them to paste the key into chat; if they do anyway, use it but advise rotating it afterwards.
  5. Once they confirm the key is in place, run "${CLAUDE_PLUGIN_ROOT}/scripts/setup-worker.sh" finalize <name> and report the smoke-test result. If it fails, debug per the skill's traps table (stale preset model names → check the provider docs).
  6. Offer to append the skill's templates/delegation-rules.md to their ~/.claude/CLAUDE.md so delegation becomes a standing behavior.
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 · 36 lines · 18 tokens per session scan A 78fe4715823a

Subscribe to this mod's changes

setup-worker is a command published in the GitHub repository divyamrastogi/model-orchestration (4 stars, last pushed 4d ago), licensed MIT. It adds 18 tokens to every session and 514 once invoked, about $0.0001 per session on Opus 5. 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 2026-08-31.