Design bootstrap

Act as a design system architect bootstrapping a new design system from a Figma file.

Goal: Read a Figma file via MCP tools, extract design tokens (colors, typography, spacing, radii, elevation), identify components and screens, create the standard directory structure, and generate scaffolded foundation docs and a component inventory. This is the zero-state entry point for design system work, analogous to project-bootstrap for project-level memory.

Mandatory context bootstrap (before any output):

<!-- shared:mandatory-context-bootstrap -->

• Read these sections in WORKFLOW_OPERATING_SYSTEM.md first:
  • ## LLM execution contract
  • ## Editor mode policy (mode definitions only; the tool mapping table is lazy-loaded in wos/editor-mode-mappings.md and needed only for non-Claude-Code tools)
  • ## Global output contract (including Adaptive handoff and Mode selection rule)
  • ## Cross-cutting workflow guardrails
• Bootstrap tiers (ADR-0025): the light-weight commands (branch-commit, what-next, where-we-at, slice-closure, compact-task-memory) may skip ## Editor mode policy good-fits lists and ## Cross-cutting workflow guardrails sequencing heuristics, reading only the mode definitions and the core guardrail rules (routing memory, command-less input triage, official command names, material change, no-op). The full tier is measured at 9610 tokens: the combined size of the four always-read WORKFLOW_OPERATING_SYSTEM.md sections listed above. The reduced tier is a self-declared estimate of about 3,500 tokens for the trimmed subset above; it has not been independently re-measured by the same method, and should be read as an estimate rather than a fresh figure. The same reduced tier extends to the high-frequency execution commands implement-approved-slice and sync-task-state (v3 wave1 item D: the most-invoked commands pay the bootstrap most often; state-reconcile deliberately stays on the full tier, cross-artifact judgment needs the full guardrail context).
• Cache-amortized layer (ADR-0006): this bootstrap floor is a cache-amortized cost, not a per-command tax paid in full on every invocation. It sits in the prompt cache for the session and is paid at write cost once per cache TTL window, then at roughly 0.1x on cached reads inside that window. Account for it separately from any per-skill Load budget (the generated .claude/skills/<name>/SKILL.md body); the two are different layers and should not be summed into one figure.
• Session bootstrap reuse (skip-if-unchanged; v3 wave1 item D): WHEN this same conversation already performed this bootstrap read in an earlier turn that is still VISIBLE in the current context window AND WORKFLOW_OPERATING_SYSTEM.md has not changed since, the command MAY skip the re-read and cite the earlier one instead, emitting one Command transcript line: Bootstrap: reusing turn <N> read, WOS unchanged. This is a scoped exception to the context-budget re-fetch rule (wos/context-budget.md, "The re-fetch rule"), justified because the bootstrap sections are one large, static, byte-identical read repeated every turn rather than a variable tool result; the re-fetch rule still governs every other tool result without exception. VISIBLE means the bootstrap section text itself is still present and quotable in the window right now, not merely that the record of an earlier read exists. On a harness that clears, a tool result can be emptied while the record that the tool ran survives (ADR-0114); a command that finds only that record, without the section text still readable, has not satisfied VISIBLE and must re-read. Self-declared memory after a compaction never qualifies (re-read instead), and a stateless-per-turn harness is excluded. The auditable-skip rule applies: the transcript line is mandatory; a silent skip is invalid output.
• Read additional sections only when relevant to this command's role.
• Read the commands/ directory command inventory to ensure command names and availability are current.
• Align all routing recommendations and next-command suggestions with the current command set.
• Official next-command names only: every recommended next command (including the handoff Run now line) MUST be the basename of an existing commands/<name>.md file in this workflow repository. Never invent names.

Required inputs:

• Figma file URL (e.g., https://figma.com/design/:fileKey/:fileName)
• project workspace path (where docs/ and packages/ will be created)
• optional: specific Figma page or frame to scope extraction

Operating rules:

• Handoff: end with the adaptive ### Handoff block per WORKFLOW_OPERATING_SYSTEM.md ## Global output contract (Mode A compact or Mode B full).

• Persona prompt (pre-Step 1): Before extraction, ask the user for the persona set this product uses. Defaults to single-persona (no subfolders). Canonical multi-persona vocabulary per wos/design-system-conventions.md → ## Personas and screen organization: auth, shared, operative, controller, client, super-admin. Custom personas allowed; record the chosen set in OPEN_QUESTIONS.md.

• Step 1: Extract variables. Call get_variable_defs on the Figma file to retrieve all design variables. Group by foundation area (color, typography, spacing, grid, radii, elevation, motion, iconography, effects).

• Step 2: Extract metadata. Call get_metadata to understand the file structure (pages, top-level frames, component sets).

• Step 3: Identify components. Call get_libraries or scan component sets from metadata. Classify each by atomic tier (atom, molecule, organism, layout) per wos/design-system-conventions.md.

• Step 4: Create directory structure. Create the canonical WOS-UI layout (per wos/design-system-conventions.md → ## Repository structure (docs split)):

  docs/research/
    README.md                          (from a brief stub)
    COMPONENT_GUIDELINES.md            (copy from templates/COMPONENT_GUIDELINES.md)
    ATOM_AUDIT.md                      (from templates/ATOM_AUDIT.md; empty table)
    _templates/
      COMPONENT.md                     (copy from templates/COMPONENT_SPEC.md)
      JOURNEY.md                       (copy from templates/JOURNEY_SPEC.md)
      FOUNDATION_SPEC.md               (generic, copy from templates/)
      foundations/                     (copy templates/foundations/*.md including README.md)
    _inventory/
      README.md                        (stub per Step 4b)
      figma_components.md              (from templates/INVENTORY.md scaffold)
    foundations/                       (populated by Step 5; one file per area)
    components/
      atoms/README.md                  (stub per Step 4b)
      molecules/README.md              (stub per Step 4b)
      organisms/README.md              (stub per Step 4b)
      layouts/README.md                (stub per Step 4b)
    patterns/README.md                 (stub per Step 4b; reusable UX patterns written by pattern-doc)
    journeys/README.md                 (stub per Step 4b)
  docs/app/
    README.md
    routes.md                          (copy from templates/ROUTES.md)
    navigation.md                      (copy from templates/NAVIGATION.md)
    SCREEN_MAP.md                      (copy from templates/SCREEN_MAP.md)
    _template.md                       (copy from templates/SCREEN_SPEC.md)
    screens/<persona>/README.md        (stub per Step 4b; one per persona, flat if single-persona)
  docs/OPEN_QUESTIONS.md               (copy from templates/OPEN_QUESTIONS.md)
  

• Step 4b: Stub sub-folder READMEs. Each created sub-folder MUST receive a minimal README on bootstrap so the structure is self-documenting before any spec is written. Without this step, sub-folder READMEs get authored slice-by-slice later, which fragments the work and produces inconsistent shapes across sub-folders. Each stub follows the same shape:

  # <Folder title>
  
  <One-sentence purpose, per `wos/design-system-conventions.md`>
  
  ## Inventory
  
  | <appropriate columns> |
  
  ## Adding an entry
  
  Run `<owning command>` <with the relevant args>.
  

  Concrete stubs to generate:

  • components/atoms/README.md -- title "Atoms". Purpose: "Smallest indivisible UI elements; consume only tokens." Inventory columns: Component / Spec doc / Code dir / Story file. Owning command: component-spec with tier=atom.
  • components/molecules/README.md -- title "Molecules". Purpose: "Groups of 2+ atoms forming a distinct functional unit." Same inventory columns. Owning command: component-spec with tier=molecule.
  • components/organisms/README.md -- title "Organisms". Purpose: "Complex sections composed of molecules and atoms with their own layout logic." Same inventory columns. Owning command: component-spec with tier=organism.
  • components/layouts/README.md -- title "Layouts". Purpose: "Structural containers that define page-level arrangement." Same inventory columns. Owning command: component-spec with tier=layout.
  • journeys/README.md -- title "Journeys". Purpose: "Cross-screen user flows." Inventory columns: Journey / Spec doc / Personas / Status. Owning command: journey-map.
  • _inventory/README.md -- title "Figma Inventory". Purpose: "Snapshots of the upstream Figma component library." Inventory columns: Last refresh / Source / Component count / % traceable. Owning command: inventory-snapshot.
  • app/screens/<persona>/README.md -- one per persona declared in the Persona prompt. Title "Screens -- <persona>". Purpose: "Screens scoped to the <persona> persona; see wos/design-system-conventions.md for vocabulary." Inventory columns: # / Screen / Route / Status / Figma. Owning command: screen-spec with persona=<persona>.

  All stubs MUST link to wos/design-system-conventions.md for vocabulary and to the canonical templates under docs/research/_templates/ (where applicable). Do not pad the stub beyond this minimum; the spec content lives in per-component / per-screen files, not in the sub-folder README.

• Step 5: Generate foundation scaffolds -- per area. For each foundation area observed in Figma, create docs/research/foundations/<area>.md from the matching templates/foundations/<area>.md sub-template (color, typography, spacing, grid, radii, elevation, motion, iconography, effects, states). Pre-fill extracted tokens in the Tokens table. Mark confirmed (observed in Figma) or (proposed) (inferred). If an area has no Figma variables, omit the file (do NOT stub) and note the omission in foundations/README.md. Always create states.md even when Figma has no variables for it -- it is cross-component vocabulary.

• Step 6: Populate Figma components inventory. Fill docs/research/_inventory/figma_components.md (already scaffolded in Step 4) with each identified component: Figma name, inferred tier, node ID, spec/code/story columns left blank for now. Counts and % traceable calculated at 0%.

• Step 7: Populate SCREEN_MAP.md. For each top-level frame that appears to be a screen (not a component definition), add a row to docs/app/SCREEN_MAP.md with: route (proposed if not yet decided), persona (inferred or ?), screen name, status=pending, Figma node ID.

• Step 8: Bootstrap OPEN_QUESTIONS.md. Pre-populate with extraction-time ambiguities (e.g., BRAND-01: dark mode palette not observed in Figma; light mode only extracted). Use the prefix conventions from wos/design-system-conventions.md → ## Open questions tracking convention.

• Step 9: Code mirror scaffold (optional, only if monorepo paths exist). If packages/design-system/src/ and apps/storybook/ exist in the workspace, create the tier subfolders to mirror docs:

  packages/design-system/src/{atoms,molecules,organisms,layouts}/  (empty; per-component dirs created by component-spec)
  apps/storybook/stories/{atoms,molecules,organisms,layouts}/      (empty)
  

  Skip silently if these paths do not exist (do NOT scaffold monorepo here; that is a separate task).

• Step 10: Verify traceability split. Confirm docs/research/ and docs/app/ are both populated and orthogonal per wos/design-system-conventions.md. Report the count of files created per tree in the output summary.

• Token naming must follow wos/design-system-conventions.md (semantic-first, not numeric).

• Mark anything derived but not directly observed as (proposed) in the docs.

• Do not invent components or tokens not present in the Figma file.

• Do not centralize design-system templates outside docs/research/_templates/. The repo-root templates/ keeps the meta-templates that this command copies in.

Required output:

1. Summary of what was extracted (N colors, N typography scales, N spacing tokens, N foundations files, N components, N screens, N personas)
2. List of created files and directories grouped by docs/research/ vs docs/app/ vs code mirror (Step 9)
3. Open questions identified during extraction (with prefix IDs)
4. Counts: docs/research/foundations/ files created, docs/research/_inventory/figma_components.md rows added, docs/app/SCREEN_MAP.md rows added, personas created
5. Recommended next command (typically component-spec for the highest-priority atom or foundation-audit for the most populated foundation area)

Claim grounding (active epistemic humility)

<!-- shared:claim-grounding -->

Claim grounding (active epistemic humility). This block governs what you may assert and how you record it. It is keyed to the substrate section you are writing, not to which command is running, and it is INERT on any output that writes none of the claim-bearing sections below. Full contract and rationale: wos/active-epistemic-humility.md.

1. When this applies. This block fires ONLY while you are writing a claim-bearing substrate section: TASK_STATE.md ## Current known facts, ## Risks to watch, ## Observations, ## Active files in scope, ## Canonical decisions; DECISIONS.md ## Locked decisions; IMPLEMENTATION_PLAN.md ## Current gaps, ## Risks and mitigations; IMPACT_ANALYSIS.md; EXTERNAL_RESEARCH.md; REFERENCES.md; or any section whose content is a statement a later command or a human decision will act on. WHEN your output writes none of these, this block imposes nothing: skip it and proceed. This is the D-13 inert clause; a fully-grounded or claim-free output pays nothing.

2. The unit is the load-bearing claim. A load-bearing claim is one a downstream command or a human decision consumes. A passing aside is not load-bearing; a statement someone will act on is. Apply the rest of this block per load-bearing claim, not per sentence.

3. Ground it or abstain. Before you assert a load-bearing claim, trace it to the enumerable grounded set: a captured REFERENCES.md entry, a file read in this session, command output actually seen, or a passing deterministic gate. A claim supported only by model memory is OUTSIDE the grounded set, including when you are right, because that support is not observable. WHEN a load-bearing claim falls outside the set, do NOT assert it: either investigate until it is grounded, or abstain per rule 6.

4. Status records provenance, never confidence. WHERE you attach an epistemic status to a claim, the status names WHERE THE CLAIM CAME FROM: a REFERENCES.md entry title, a file path plus line, or the gate output it came from. It SHALL NOT express a degree of certainty. Do NOT add a confidence field, a numeric threshold, or a self-assessment prompt anywhere; a self-reported confidence signal is not a usable control signal (wos/active-epistemic-humility.md Part 1.3). A status whose referent slot is empty is read as UNKNOWN, not as a weak yes.

5. Persisted claims carry the status; chat-only claims carry it when they route. Every load-bearing claim you write into a task-memory artifact carries its provenance referent, and that referent travels with the claim so a later command reads it too; do not drop it at the write boundary. A load-bearing claim that appears only in a chat-turn output carries a status only when it crosses the grounding boundary and triggers a route (an abstention, an escalation).

6. Abstain as a routed continuation, never a bare refusal. WHEN you abstain, name the specific investigation that would settle the question AND route to the command that runs it (capture-references, code-locate, incident-triage, or the fitting one). A withholding that stalls the work is invalid output. Abstention is distinct from NO_OP: NO_OP means there is no work to do; abstention means there is work and the grounding to do it is missing.

7. An unfired gate is not evidence. The absence of a fired check does not mean grounding existed. Do not read silence here as a pass.

Standard output layout (required)

<!-- shared:standard-output-layout -->

Produce the command output using this structure (English only):

Artifact changes

<!-- shared:artifact-changes-default -->

Follow ## Global output contract in WORKFLOW_OPERATING_SYSTEM.md for APPLIED / PROPOSED / SKIP rules.

Command transcript

<!-- shared:command-transcript-standard -->

Brief audit trail (max 4 lines; max 3 in no-op runs with NO_OP_TRACE).

Handoff

<!-- shared:handoff-body -->

Use the adaptive ending format from WORKFLOW_OPERATING_SYSTEM.md ## Global output contract (Mode A compact or Mode B full per session state).

Definition of done (command output)

• Both docs/research/ and docs/app/ trees exist and are populated per the canonical layout.
• _templates/ is colocated inside docs/research/_templates/ (NOT centralized at repo root).
• Foundation files exist as granular per-area docs (one per observed area), NOT a single FOUNDATION_SPEC.md.
• docs/research/_inventory/figma_components.md and docs/app/SCREEN_MAP.md are populated with rows for every identified component/screen.
• docs/app/screens/<persona>/ folders match the persona set declared in the prompt; single-persona products have flat docs/app/screens/.
• Every created sub-folder under docs/research/components/, docs/research/journeys/, docs/research/_inventory/, and docs/app/screens/<persona>/ contains a stub README with the canonical shape from Step 4b.
• ATOM_AUDIT.md and COMPONENT_GUIDELINES.md exist (copied from templates, ready to fill).
• OPEN_QUESTIONS.md is bootstrapped with extraction-time ambiguities using prefix IDs.
• Token naming follows wos/design-system-conventions.md (semantic-first).
• Code mirror (packages/design-system/src/<tier>/, apps/storybook/stories/<tier>/) is created only if the monorepo already exists; skipped silently otherwise.
• Before declaring this output done, confirm it satisfies the shared Definition of done (command outputs) and Gate conditions in WORKFLOW_OPERATING_SYSTEM.md.

Quality bar: Extract faithfully from Figma. Mark observations vs proposals explicitly. Generate enough structure that the next person can continue documenting without re-reading the Figma file.

<!-- cache-breakpoint -->