Build A Brand

Build a Brand

Tools below are Pika MCP tools, named bare — call each under whatever prefix your session exposes for the Pika MCP.

Take any input — an idea, a website, a list of reference brands, product photos, or an existing brand to refresh — and produce a usable brand identity.

This is a standalone brand-building workflow focused on strategy, identity, design, and copy. The default output is a quick brand deliverable that can be finished in minutes. The full 14-16-page brand guidelines PDF is still available, but only when the user explicitly asks for full depth or chooses to go deeper after the quick pass.

Reference files

Load these only when the relevant phase starts:

• references/brand-directions.md — strategy angle structure and differentiated positioning boards.
• references/brand-identity.md — logo pipeline, symbol rules, concept lanes, and board-copy budgets.
• references/brand-guidelines.md — full guidelines layout, render contract, page structure, and QA rules.
• references/brand-md-template.md — required machine-readable brand.md structure.

Execution Model — Local First, Cloud Only Where Needed

This skill is local-first. Keep deterministic production work on the user's machine:

• Local: workspace setup, downloaded fonts, generated image files after download, image compression, transparent-background cleanup, 16x16 favicon tests, HTML/CSS page builds, PDF rendering, PNG/JPG QA screenshots, logo asset assembly, token/prompt files, and final zip packaging.
• Cloud: paid generation only (gpt-image-2 for symbols, mood images, photography, illustration, and ambient textures) and URL/source research when the brief requires it.

Do not use a cloud PDF renderer or upload PDFs by default. Save quick PDFs, board PDFs, guidelines PDFs, and brand-kit zips to ~/Desktop on Mac, or the project working directory when Desktop is unavailable. Only upload/share via CDN if the user explicitly asks for a hosted file.

PDF asset hygiene: never embed full-size gpt-image-2 PNGs directly in board or guidelines PDFs. This previously produced huge PDFs and ASSET_FETCH_TIMEOUT failures when renderers tried to fetch multi-MB CDN images. Download every generated image, down-raster/downsample it to the slot size, and save a local JPEG target of about 85-180KB before using it in render HTML. Keep transparent PNG only for logo/symbol source assets and favicon/logo export, not for photographic page imagery. Never put a CSS `filter:`/`backdrop-filter:` (warm grade, sepia, saturate, brightness, blur) on a photo or `background-image` in the render HTML — it forces Chromium to rasterize that element to a full-resolution lossless bitmap in the PDF, which cancels out the JPEG step above and re-bloats the file (a warm-graded mood-board page alone took a deck to 70 MB). Bake any color grade into the JPEG itself (in the gpt-image-2 prompt or when you img.save(..., 'JPEG', ...)); drop-shadow() on a transparent logo PNG is the only filter that's safe.

Cost transparency gate

Before any paid MCP call, call identity_balance({verbose: true}) once. Surface the current balance, recent burn rate, and remaining runway, then gate the selected depth with this exact message:

Estimated cost: about 100-300 credits (~$1-$3) for quick brand, or about 900-1,500 credits (~$9-$15) for full brand book. These ranges cover gpt-image-2 symbol / mood / touchpoint generation plus one targeted regeneration for failed image QA. Quick brand is usually below $5; full brand book exceeds $5, so Reply proceed to continue or cancel to stop.

Do not call any paid MCP tool until the user replies proceed. If the user replies cancel, stop without generating. For non-interactive --quick, --full, or --config callers, require cost_ack=proceed in the config or invocation metadata; if it is absent, stop with the estimate instead of spending credits.

Depth Modes

Choose the depth before any long-running generation or render work:

1. Quick brand (default) — use for first-time founder-kit runs, --quick, creator tutorial flows, batch/subagent calls, or any user who asks for a brand without explicitly asking for a full brand book. Target: 8-12 minutes after usable input, excluding user reply time; no more than 4 generated images unless QA fails; one quick PDF render plus one QA rerender when needed.
2. Full brand book (opt-in) — use only when the user asks for "full", "brand book", "brand guidelines", "14-16-page", "15-page", --full, or when they approve the quick path's "go deeper" upgrade. Target: 25-45 minutes after choices are locked, excluding user reply time.

The quick brand is not a partial or failed full guidelines deck. It is a deliberately smaller deliverable with its own completion criteria. If the selected mode is full, all full-guidelines completion gates still apply.

Mode Selection

Stage 0 — Intake (empty-args menu)

If invoked with no input (no idea, no URL, no photos, no reference brands, and no relevant prior context in the conversation), print this menu verbatim as your full response and stop. Do not call any tool. Wait for the user's next message.

What are we branding? Paste any of: - An idea / description — e.g. "a streetwear label for cat people" - A website URL — e.g. "rebrand my existing site at example.com" - Product photos — drop them in the chat - Reference brands — e.g. "I want something that feels like Aesop + Patagonia" - An existing brand to refresh — name + what's working / what isn't I'll start with a quick brand by default. Say full brand book if you want the deeper 14-16-page guidelines path.

If the user already dropped one of the above, skip the menu and proceed to the selected mode: Quick Step 1 for the default quick brand path, or Step 1 below when full brand book mode is selected.

Stage 0.5 — Quick brand and non-interactive fast lane

Use quick brand mode when the caller passes --quick, omits an explicit full-depth request, or states they are running from CI, a subagent, a batch job, a tutorial recording, or any other non-interactive harness.

Use full brand book mode only when the caller passes --full, sets mode: "full" in config, explicitly asks for full guidelines / brand book depth, or approves the post-quick "go deeper" upgrade.

This section has precedence over the interactive ask/wait instructions below. When quick mode applies, use the quick brand workflow below and do not fall through to the multi-turn full-workflow intake unless a required input is truly missing.

• --config <path> points to a JSON file that pre-bakes intake answers: input, photos, reference_brands, audience, positioning, assets_to_keep, references, mode, chosen_direction, chosen_identity, and export_kit.
• --quick means quick brand mode. Use model judgment for all confirmation gates. Ask only if the original input is missing entirely; otherwise infer reasonable defaults, choose the strongest strategy direction and identity option, and continue.
• --full means full brand book mode. Still use model judgment for confirmation gates in non-interactive runs, but do not cut the required 14-16-page guidelines.
• For --quick or --config, do not stop for confirmation at the deliverable preview, strategy-direction choice, identity-option choice, or brand-kit export gate. Record the assumption inline, then proceed.
• If a required asset is unavailable and cannot be inferred from the input, stop once with a single compact missing-fields list instead of starting a multi-turn Q&A loop.
• In quick mode, deliver the quick brand package defined below; do not describe it as a condensed full guidelines deck.
• In full mode, do not deliver a condensed or partial brand output just because the caller is non-interactive or the run is short on wall-clock time. A