offload-to-locallisted

Use when PLANNING or decomposing ANY multi-step task, and whenever your plan involves reading, searching, summarizing, transforming, drafting, extracting, or editing across files — regardless of whether cost is mentioned. Also when you're about to do delegatable/bulk/mechanical work on a paid cloud model, or you notice you're about to burn cloud tokens on work a smaller model could do. It routes delegatable work to a free local model — by DEFAULT to the broad `operator` role, escalating to a specialist role only when needed — instead of spending cloud tokens. Explicit triggers: starting a multi-step plan (especially under a budget constraint / Credit Efficient Mode), "save cost/tokens", "do this locally", "offload this", "use the local model for this". ALSO use before ANY delegation once cost or free/local agents have been mentioned, and whenever you are about to call the Agent/Task tool hoping it will be cheap — it cannot reach a free model and defaults to a full-price subagent, so read this for the routes t
haiggoh/free-agents · ★ 1 · AI & Automation · score 67

Install: claude install-skill haiggoh/free-agents

# offload-to-local — send the legwork to a free local model The biggest cost saving in Claude Code is not a cheaper cloud model — it's **$0 local compute**. When your main session runs a capable cloud model (Opus/Sonnet), keep that model for judgment and **delegate the bulk/mechanical legwork to a local model** via dispatch. This applies to anyone who wants to save cost, with or without a spending cap. ## ⛔ READ THIS FIRST — the built-in `Agent` tool CANNOT reach a free model This is the single most common way this skill gets "used" without saving a cent, so it is stated before anything else. **Claude Code's `Agent` (Task) tool cannot dispatch to a free agent. Ever.** Its `model` parameter accepts only first-party paid tiers — `sonnet`, `opus`, `haiku`, `fable`. There is no value that routes to a local MLX model or to a free cloud API, and the org allowlist blocks adding one. So: | What you do | What actually runs | Cost | |---|---|---| | `Agent(...)` with **no** `model` | the **default subagent model** (usually the same tier as you) | **full price** | | `Agent(..., model="haiku")` | Haiku, on the paid gateway | **cheaper, still paid** | | `Agent(..., model="qwen…"/"local"/"free")` | rejected — not a valid enum value | n/a | | **`curl` / the dispatch scripts below** | a free local or free-API model | **$0** | **Omitting `model` is the trap.** It reads like "let the harness pick something cheap"; it actually means "run a full-price subagent." A dispatch that went through