# Brainstorm Build Lite > The lightest "brainstorm then build". Use when the user wants a quick, cost-aware build that offloads the simpler or mechanical parts to a cheaper model and doesn't need Fable's creative brainstorm — "just design and build it, cheap where it can be". For a straight all-Opus plan→build (no Sonnet offload) use brainstorm-build-mid; for a Fable-led creative design plus the full ceremony use brainstorm-build-prime. NOT for tiny one-line edits (just do them) or pure design/no-build work. - Skill: `bodish-coder/brainstorm-build-lite` (Agent Skill) - Install (CLI): `npx skillmds@latest add bodish-coder/brainstorm-build-lite` - Raw SKILL.md: https://api.skillmd.com/api/skills/bodish-coder/brainstorm-build-lite/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - Author: bodish-coder (https://skillmd.com/u/bodish-coder) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/bodish-coder/brainstorm-build-lite --- # Build (Opus design + build) + Sonnet (simple tasks) Two models, no Fable. **Opus** does the design thinking and the complex/core implementation; **Sonnet** offloads the simple, mechanical tasks so you don't pay Opus rates for boilerplate. A skill can't change the main session's model, so the split is done with **subagents** carrying explicit `model` overrides. You (the main session) orchestrate only — spawn the agents, route by complexity, relay the result. ## Craft — read this first Read **`PRACTICE.md`** at the plugin root (beside `PLATFORMS.md`). It is skillator's process canon — the superpowers process skills merged in, in skillator's own words — so none of them is chained in front of this one: classify the request spike/bounded/architectural and say which out loud; one question at a time; the plan-grade TASKS shape a build agent can work from cold; design self-review; the test-first law; fresh-evidence verification; debugging (§7); branch lifecycle (§8). Everything below assumes it. ## Phase 1 — Opus designs Dispatch one subagent, `model: "opus"`, `subagent_type: "general-purpose"`. Give it the user's task text and the repo context. Ask it to commit to an approach and **write** an implementation-ready design **with each task tagged by complexity** to a file, returning only that path plus a two-line summary: ``` CHOSEN: DESIGN: CONSTRAINTS: TASKS: VERIFICATION: ``` **The design file — one file, no ceremony.** Write it to the scratchpad as `design--.md`. It is not a session record and this tier keeps none; it exists because `practice/task-loop.md` and every template in `practice/prompts.md` take a `` path (`taskwork.sh brief `), and because a Sonnet agent working one task needs the contracts on disk rather than re-summarised into its prompt. Confirm the path in your reply. ## Phase 2 — Build (route by complexity) Work the TASKS list, routing each: - **[COMPLEX] / core → `model: "opus"`** subagent. - **[SIMPLE] / mechanical → `model: "sonnet"`** subagent. Each agent gets the **design file path** and its own task block — never the design pasted into the prompt, never this session's history (PRACTICE.md §4; procedure in `practice/task-loop.md`, prompt text in `practice/prompts.md`). It builds exactly that and stays inside its declared files. Independent tasks can run in parallel (git worktree if they'd touch the same tree). If the design hits a real blocker only the user can resolve, stop and surface it — don't guess. ## Phase 3 — Verify & relay Run the VERIFICATION step (Opus agent or main session) through PRACTICE.md §5: identify the command, run it in full now, read the whole output and exit code, and only then claim — with the evidence attached. Never from an earlier run. On failure, fix it ([SIMPLE]→Sonnet, [COMPLEX]→Opus) and re-run. Then relay concisely: the approach, what was built (files), and the verification result. ## Workflow mode — wide builds When the design returns **4+ independent tasks**, or the work is a sweep (migration, audit, codemod), or the user asked for it, run Phases 1-3 as a **single deterministic workflow script** instead of hand-dispatched subagents: design → one build agent per task → verify each → loop the failures. Read **`WORKFLOW.md`** (beside the installed skills, or at the repo/plugin root) for the criteria, host table, and a ready script. Write the design file before the call and pass its path in `args`, exactly as Phase 2 would. The [SIMPLE]/[COMPLEX] routing survives the switch — it becomes the schema's `complexity` field, and the build stage picks the model per task: `model: t.complexity === 'SIMPLE' ? 'sonnet' : 'opus'`. Same rule, one call. For 1-3 sequential tasks, stay with plain dispatch. ## Rules - **Route by tag, not vibe.** [SIMPLE]→Sonnet, [COMPLEX]→Opus, design→Opus. Don't send boilerplate to Opus (waste) or core logic to Sonnet. - **Both build phases are subagents.** Don't design or code in the main session. - **Autonomous but honest.** No confirmation gate; but if a phase fails or an agent returns nothing, say so and stop rather than continuing on a missing piece. - **Handoff before any context loss.** This tier runs no `/compact` or `/clear` of its own, but if you or the user are about to run either, **first run the `handoff` skill** — never compact/clear without a verified handoff. - **Pass the design by path, never by paste.** Pasting it parks the whole design in your context for the rest of the session (`practice/task-loop.md`, Context hygiene), and the canon's templates take a `` path anyway. - **Minimal ceremony, real canon.** The scratchpad design file is the only artifact this tier writes — no session record, no Outcome section, no rework loop, and no §6 whole-branch reviewer — that end-of-build `deep`-tier pass is prime's ceremony, and this tier seats no `deep` model. It ships on §5 evidence plus §4's per-task review. PRACTICE.md §§1-5, and §6's receiving-review discipline, still apply. - Need Fable's creative brainstorm, a handoff checkpoint, or a kept session record? Use brainstorm-build-prime. Want all-Opus (no Sonnet offload)? Use -mid. ## Other hosts Opus/Sonnet and the Agent tool above are the **Claude Code** defaults. On Cursor, Codex, Antigravity, Pi, or Prime Agent, map them through `PLATFORMS.md` (beside the installed skills, or at the repo/plugin root): design + [COMPLEX] → **build tier**, [SIMPLE] → **cheap tier**, dispatched with that host's delegate mechanism. Where the host has no delegation, run the phases **sequentially in one session**, switching model between them and writing each phase's output to disk before the switch — the file is the handover, not chat history.