Wix Headless Replatform
WixOfficialPublished by Wix
- Skill
- Workflow
- Software development
- MIT
- Low risk
- Has scripts
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
The wix-headless-replatform skill acts as a central supervisor for migrating or cloning a public website's home page into a Wix Headless environment. It enforces a structured, multi-stage process that includes browser-based extraction, build planning, implementation, and visual verification. The skill ensures visual fidelity, SEO preservation, and structural integrity by utilizing a frozen manifest approach, where every stage produces durable artifacts that inform the next. It supports both standalone cloning and integrated migration modes, with specific protocols for handling gaps and ensuring the final output is a functional Wix CLI project.
What it does
- Clones or migrates a public website home page to a Wix Headless frontend.
- Manages the end-to-end workflow through distinct stages: context, evidence, implementation, and verification.
- Uses browser-backed extraction to capture visual and structural data.
- Enforces strict visual fidelity and SEO preservation standards.
- Supports one-click automation for migration handoffs.
- Generates a final report and terminal receipt for migration completion.
When to use it
- When you need to clone a public website home page into a new Wix Headless project.
- When performing a migration of an existing site to a Wix Headless architecture.
- When you require a structured, verifiable process for rebuilding a frontend with high visual fidelity.
- When working within a migration handoff that provides specific project and data-binding intent.
Advantages
- Ensures consistent results through a rigid, stage-based supervisor protocol.
- Maintains high visual and functional fidelity to the source website.
- Provides a clear path for recovery and gap resolution during the build process.
- Automates the transition from extraction to a runnable Wix CLI scaffold.
Usage examples
- “Clone the home page of a website into a new Wix Headless project.”
- The agent initiates the standalone clone workflow, resolving the source URL and proceeding through the context and evidence stages to create a new Wix CLI project.
- “Migrate this website using the provided handoff file.”
- The agent enters migration mode, reads the handoff configuration, and continues the build process within the existing migration-owned project structure.
Install
Download the skill as a ZIP:
https://autogineer.com/agent-skills/wix/wix-headless-replatform-p46x/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip wix-headless-replatform.zip -d .agents/skills
Documentation by Wix
wix-headless-replatform
Role
Drive a close-enough, visually polished Wix Headless frontend from a source website.
The implemented scope is one resolved home page and the default output is
projects/<project-name>, derived from the hostname. Use Wix CLI Headless, Astro,
TypeScript, and Tailwind; use React only for complex interactive islands.
Reject additional-page and explicit-URL scope with an actionable explanation until a later
approved workflow defines page resolution and shared evidence across routes.
This entry skill is the supervisor. It decides the next workflow stage and loads only the resource for that stage. Do not preload every resource or reference: stage-specific rules are deliberately demand-loaded.
Two modes exist:
Standalone clone: source URL to a new, real Wix CLI Headless project.
Migration phase:
--handoff <migration-project>/website/handoff.jsoncontinues on the migration-owned destination. Never omit--handofffor an integrated migration: the standalone command creates a second site.
Non-negotiable invariants
Treat the requested public branded homepage as authorized; apply the boundaries in
references/safety.mdwithout asking a separate routine authorization question.Browser-backed extraction is mandatory. Do not substitute HTML-only extraction.
Resolve the home page before capture. Build only from a validated frozen extraction manifest; raw observations are never builder inputs.
A standalone output must be a successful Wix CLI scaffold with valid Wix identity and runnable
devandbuildscripts. Never hand-create a Wix-shaped substitute.In migration mode, reuse the handoff's site/project and route/data-binding intent; do not create or retarget a second destination.
Preserve source structure, first-viewport evidence, chrome hierarchy, source assets, fonts, core interactions, SEO, internal URL behavior, and branded fallback behavior.
A build is not completion. Completion requires the accepted post-build visual-review and gap-loop evidence described by
rp-qa-gap-loop, plus optional facelift acceptance only when the user explicitly selected facelift mode.
Supervisor protocol
On every new or resumed run:
Resolve the source URL, output directory, and mode. In migration mode read
website/handoff.jsonanddocs/site-clone/frontend-automation-state.jsonfirst.Select the first incomplete stage from the table below and read only that resource's
SKILL.mdplus the references it explicitly needs.Execute that stage until it writes its required durable artifact, or records a specific genuine blocker with evidence and an exact unblock action.
Re-select the stage after every boundary. Do not say that a required next step “needs to happen” and end the run; load its resource and perform it.
| Stage | Load only when | Resource | Durable outcome |
|---|---|---|---|
| Context | mode, scope, project, or browser environment is unresolved | resources/rp-project-context/SKILL.md | valid context, approval record, runnable project/plan state, or blocker |
| Evidence | frozen home-page extraction is absent or stale | resources/rp-source-evidence/SKILL.md | validated extraction/latest.json and frozen manifest, or global blocker |
| Implementation | frozen build plan exists and home UI is incomplete | resources/rp-ui-implementation/SKILL.md | manifest-hash-aware runnable home implementation |
| Verify and improve | build, QA, comparison, review, fix, report, release, or receipt is incomplete | resources/rp-qa-gap-loop/SKILL.md | accepted result, done_with_gaps, or global blocker |
| Optional facelift | normal clone accepted and automation state says facelift is requested | resources/rp-facelift/SKILL.md | accepted facelift review, or blocker |
When frontendPhase.allowedNow is plan, complete only the plan-eligible context and
evidence work and leave a resumable plan-state record. When it is build, continue through
implementation and verification against the migration-owned project.
One-click continuation
A migration handoff declaring automationMode: "one_click" authorizes routine scope and
frontend checkpoints. Record those approvals in
docs/site-clone/frontend-automation-state.json; do not pause to ask whether to proceed
with a required stage, screenshot review, or authorized gap-fix pass.
An explicit facelift request may be carried from 1-click intake, but it remains gated until
normal clone acceptance; 1-click never implies facelift by itself.
In a one-click build run, keep selecting and executing stages until one of these durable outcomes exists:
website/completion.jsonis valid and the latest visual review is complete with no open critical/high identity finding; orfive visual gap cycles have been completed and
website/completion.jsonrecords the truthful terminal statusdone_with_gapswith the final reviewed evidence; ora genuine blocker is recorded with owner, evidence, attempted recovery, and exact next authority/action needed.
An owned local extraction gap is not a genuine blocker. Continue unrelated work, use no more
than two distinct targeted recovery attempts, then build the affected dependency closure
provisionally and surface post-assembly choices in final-report.json and .md.
Before classifying a shell, browser, or network failure as a blocker, use the
repo-local Wix/Yarn toolchain PATH ordering and follow the recovery order in
resources/rp-project-context/SKILL.md: aligned-shell retry, then the
required browser/network escalation and retry, then blocker only if it still
fails or approval is pending. After every gap iteration, immediately return to
the stage table; never treat a first build, a reported plan, or backend import
as frontend completion.
In migration mode the build must also leave the final route manifest that old-link
preservation consumes (output-contract.md, "Migration terminal receipt"). A built site is
not proof that old URLs work; redirect application and public verification stay pending
until the cutover step runs them.
Commands
Run node scripts/site-clone.mjs <source-url> for standalone mode, or
node scripts/site-clone.mjs <source-url> --handoff <path> for migration mode. The active
resource gives the exact commands and inputs for its stage.
References
Resources load their own references. The shared contracts are kept under references/:
safety.md, workflow.md, extraction-schema.md, output-contract.md,
qa-checklist.md, and the interaction/layout/control/normalization/gap contracts.
Files
- .publish-manifest.json228 B
- ATTRIBUTION.mdAdded by AutoGineer833 B
- LICENSELicence file added by AutoGineer1.0 KiB
- SKILL.md6.7 KiB
- VERSION6 B
- agents/openai.yaml259 B
- references/control-and-visual-fidelity-contract.md2.1 KiB
- references/extraction-schema.md8.6 KiB
- references/interaction-runtime-contract.md5.5 KiB
- references/layout-blueprint-contract.md2.1 KiB
- references/output-contract.md5.0 KiB
- references/post-build-gap-contract.md5.5 KiB
- references/qa-checklist.md4.7 KiB
- references/safety.md1.0 KiB
- references/ui-normalization-contract.md2.8 KiB
- references/workflow.md8.5 KiB
- registry/README.md958 B
- registry/approvals/button/shadcn-default-ac60ef5.json980 B
- registry/items/button/shadcn-default-ac60ef5/button.tsxScript1.9 KiB
- registry/items/button/shadcn-default-ac60ef5/capabilities.json2.7 KiB
- registry/items/button/shadcn-default-ac60ef5/contract.md2.1 KiB
- registry/licenses/shadcn-ui-ac60ef5-MIT.txt1.0 KiB
- registry/registry.json1.4 KiB
- schemas/build-plan.schema.json628 B
- schemas/candidate.schema.json949 B
- schemas/common.schema.json1.2 KiB
- schemas/decision-patch.schema.json621 B
- schemas/decision-request.schema.json825 B
- schemas/extraction-gap.schema.json1.6 KiB
- schemas/extraction-manifest.schema.json1.6 KiB
- schemas/final-report.schema.json744 B
- schemas/foundation.schema.json950 B
- schemas/metadata.schema.json801 B
- schemas/observation-packet.schema.json661 B
- schemas/page-capture.schema.json1015 B
- schemas/page-resolution.schema.json1.4 KiB
- schemas/shared-chrome.schema.json940 B
- schemas/unit.schema.json1.8 KiB
- scripts/assemble-extraction.mjsScript36.1 KiB
- scripts/benchmark-extraction.mjsScript9.1 KiB
- scripts/browser-extraction-preflight.mjsScript1.2 KiB
- scripts/build-extracted-homepage.mjsScript17.1 KiB
- scripts/clear-wix-starter.mjsScript2.3 KiB
- scripts/create-wix-project.mjsScript17.7 KiB
- scripts/discover-urls.mjsScript4.4 KiB
- scripts/extract-assets.mjsScript4.6 KiB
- scripts/extract-design-system.mjsScript23.9 KiB
- scripts/extract-interactions.mjsScript83.5 KiB
- scripts/extract-page.mjsScript118.0 KiB
- scripts/extract-seo.mjsScript848 B
- scripts/extract-tokens.mjsScript2.6 KiB
- scripts/finalize-extraction-report.mjsScript6.4 KiB
- scripts/finalize-gap-review.mjsScript12.2 KiB
- scripts/finalize-gap-review.test.mjsScript5.3 KiB
- scripts/finalize-migration-website.mjsScript6.2 KiB
- scripts/generate-control-state-contract.mjsScript9.6 KiB
- scripts/generate-design-md.mjsScript11.7 KiB
- scripts/generate-extraction-validators.mjsScript5.4 KiB
- scripts/generate-invoice-download-backend.mjsScript16.5 KiB
- scripts/generate-layout-blueprint.mjsScript8.7 KiB
- scripts/generate-repeater-cms.mjsScript4.9 KiB
- scripts/generate-route-plan.mjsScript4.6 KiB
- scripts/generate-scene-contract.mjsScript16.2 KiB
- scripts/generate-ui-normalization.mjsScript13.3 KiB
- scripts/generate-visual-assets.mjsScript6.3 KiB
- scripts/lib/browser-tooling.mjsScript19.9 KiB
- scripts/lib/common.mjsScript9.0 KiB
- scripts/lib/component-capabilities.mjsScript7.9 KiB
- scripts/lib/component-registry.mjsScript15.7 KiB
- scripts/lib/content-boundary.mjsScript2.8 KiB
- scripts/lib/extraction-contract.mjsScript19.1 KiB
- scripts/lib/extraction-detectors.mjsScript11.4 KiB
- scripts/lib/extraction-validators.generated.mjsScript25.2 KiB
- scripts/lib/font-contract.mjsScript10.0 KiB
- scripts/lib/frontend-automation-state.mjsScript6.6 KiB
- scripts/lib/gap-analysis.mjsScript26.1 KiB
- scripts/lib/html-extract.mjsScript5.7 KiB
- scripts/lib/interaction-bootstrap.mjsScript296 B
- scripts/lib/interaction-runtime.mjsScript8.1 KiB
- scripts/lib/interaction-timeline.mjsScript13.9 KiB
- scripts/lib/layout-blueprint/classifiers.mjsScript7.8 KiB
- scripts/lib/layout-blueprint/geometry.mjsScript1.8 KiB
- scripts/lib/layout-blueprint/vocabulary.mjsScript872 B
- scripts/lib/playwright-request-guard.cjsScript2.4 KiB
- scripts/lib/ui-normalization.css1.5 KiB
- scripts/lib/url-safety.mjsScript6.8 KiB
- scripts/optimize-svg.mjsScript895 B
- scripts/post-build-gap.mjsScript16.5 KiB
- scripts/provision-repeater-cms.mjs2.8 KiB
- scripts/record-gap-decision.mjs2.3 KiB
- scripts/replay-extraction.mjs5.3 KiB
- scripts/resolve-home-page.mjs4.0 KiB
- scripts/site-clone.mjs35.8 KiB
- scripts/validate-font-contract.mjs1.3 KiB
- scripts/verify-interactions.mjs32.1 KiB
- scripts/visual-qa.mjs20.6 KiB
references/extraction-schema.md
Extraction-first artifact contract
Schema family: 0083.1. The hand-authored JSON Schema 2020-12 files under
schemas/ are authoritative. scripts/generate-extraction-validators.mjs
mechanically generates the dependency-free runtime validator committed at
scripts/lib/extraction-validators.generated.mjs; --check rejects schema and
validator drift. Semantic and cross-artifact checks remain in
scripts/lib/extraction-contract.mjs. Product runs do not download a validator.
Scope and order
This workflow resolves and reconstructs one home page. Do not accept non-home scope or explicit additional URLs until a later approved workflow defines their resolution and shared-evidence rules.
Resolve the home page and
/destination route.Capture browser evidence across the required viewports and safe states.
Segment visible regions and assign exactly one owner.
Extract the foundation, metadata, and shared chrome.
Recursively extract section/component units.
Enforce the rendered-content boundary, then validate and freeze accepted or provisional specs.
Content-bearing fields come from browser-rendered text, never raw parent textContent.
script, style, noscript, and template descendants are non-content even when a
parent section is visible. Visible pre/code remains valid authored content and carries
visible-code provenance. A strong source-code contamination signal omits only the affected
strings, creates an owned local gap, and makes that unit provisional without blocking the
rest of the home-page workflow. The content-extraction semantics version participates in
the capture fingerprint so a capture made under older text semantics is not reused as the
same capture.
7. Derive a build plan only from the frozen manifest.
8. Build, verify, recover bounded gaps, and issue the final report.
Canonical tree
docs/site-clone/
page-resolution.spec.json # early durable resolution checkpoint
extraction/
latest.json # capture id + frozen manifest hash only
<capture-id>/
page-resolution.spec.json
page-capture.spec.json
observations/
page.json
assets.json
seo.json
tokens.json
interaction-map.json
scene-contract.json
control-state-contract.json
visual-assets.json
layout-blueprint.json
ui-normalization.json
recursive-jobs.json
packets/*.packet.json # typed, sanitized replay inputs
foundation.spec.json
metadata.spec.json
shared-chrome.spec.json
sections/<section-id>/section.spec.json
sections/<section-id>/components/<component-id>.spec.json
extraction-gaps.json
spec-index.json
extraction-manifest.json
replay-report.json # replay captures only
build/
build-plan.json
component-selection.json
registry-installation.json
section-implementation.json
qa/
visual-qa.json
font-validation.json
registry-opportunities.json
final-report.json
final-report.mdObservation files are retained inputs to deterministic projection. Their sanitized,
typed packet forms are the only replay inputs, and every packet/reference is paired
with a semantic hash in the manifest. Packet tampering therefore invalidates the
manifest before replay. Builders must not read observations or packets.
spec-index.json maps stable unit IDs to frozen spec paths;
extraction-manifest.json is the only authority for planning a build.
Frozen specs
Every foundation, metadata, shared-chrome, section, and component spec has:
schemaVersion,kind, stableid, andpageKey: "home";status: "accepted" | "provisional";source-neutral content/layout/style/behavior/reconstruction/acceptance fields appropriate to its kind;
evidence/provenance references;
direct dependency hashes;
owned
gapRefs;an immutable SHA-256 semantic hash.
An accepted spec supplies every required fact. A provisional spec supplies every reliably observed fact and references at least one local gap naming every omitted required field, assumption, dependency, and affected acceptance check. Builders may implement the reliable portion but may not invent exact copy, assets, links, business actions, or unobserved behavior.
Serialization and hashing
Use RFC 8785 canonical JSON and SHA-256. Semantic hashes include schema version, semantic fields, evidence references, detector versions, and direct dependency hashes. They exclude timestamps, formatting, logs, and machine-specific absolute path prefixes. A changed spec or dependency gets a new hash and invalidates stale downstream ledger entries. A hash proves identity/integrity, not confidence.
All product artifact schemas reject unknown top-level fields. Extension data is
allowed only in the explicitly named extensions object with a namespaced key.
Acceptance authority
Hard gates are deterministic and cannot be waived by the agent: schema validity, referential integrity, exact ownership or explicit ignore/gap, required provenance, safe-probe compliance, dependency closure, hash integrity, and no unresolved global blocker.
Per-detector confidence is calibrated on the approved corpus. A high band must meet at least 95% precision and may freeze automatically. Medium candidates become compact typed decision requests for the orchestrating agent. Low confidence, agent abstention, or invalid patches become gaps. The agent chooses only among the typed choices and cannot override hard gates.
Anti-stuck gaps
Each gap declares owner, local/global scope, missing fields, reason, confidence, evidence, dependency closure, acceptance impact, attempts, remaining budget, assumptions, omissions, and user decision.
A local gap never stops unrelated extraction, freeze, build, or QA.
Affected work receives at most two distinct targeted attempts. Repeating the same tactic without new evidence is rejected as no progress.
After the budget is exhausted, the affected dependency closure freezes and builds provisionally.
Only unresolved home identity, unusable page-wide capture, corrupt integrity, unisolatable safety failure, or inability to assemble/build/serve any viable shell is a global blocker.
Local gaps remaining at handoff produce
done_with_gapsand a final report; they do not produce a blocked run.
After assembly the user may accept, retry/fix, provide material, replace, omit, or leave each provisional unit unresolved. Human acceptance does not relabel an unverified unit as verified.
Safe behavior evidence
Probe only public unauthenticated pages. Allow scrolling, hover, focus, keyboard navigation, resize, bounded waits, and controls confidently classified as local presentation changes (tabs, disclosures, menus, carousel controls). Block form submission, authentication, cart/checkout, purchase/booking, uploads, account or admin state, destructive controls, navigation, mutating requests, and any unclassified click. Record blocked actions/requests and hard action/time budgets.
Evidence lifecycle
Keep evidence local until explicit cleanup. Redact authorization/cookie headers, secrets, sensitive query values, and form values before persistence. Content-hash and deduplicate downloaded assets. Full browser traces/video are opt-in diagnostic artifacts. Cap retained evidence at 250 MB, frozen specs and manifests at 10 MB, typed agent decisions at 30, agent input at 50,000 tokens, and p95 extraction at ten minutes on the approved benchmark.
Replay retained packets unless source fingerprint, required observation kind, viewport/state set, capture semantics, or packet integrity changed. Detector, schema projection, or typed-decision changes alone do not require recapture.
Registry boundary
The shadcn-format registry under registry/ contains only pinned, human-approved
runtime revisions. Candidate intake and review happen outside the published
skill under the repository's registry-candidates/ workspace; deterministic
promotion copies only approved source, capabilities, the compact agent contract,
license, and approval record into registry/.
Selection code filters hard license/accessibility/responsive/SSR/RTL/reduced-
motion/runtime requirements and the hash-pinned generic capabilities.json
axes, slots, states, compositions, and constraints. It records rejection
evidence and the resolved typed binding in build/component-selection.json.
The building agent reads that selection record and, only for a selected item,
the referenced contract.md. It does not scan the registry or read capability
JSON, approvals, licenses, source provenance, candidate reviews, or live
documentation. When nothing qualifies, use the bounded custom builder.
Requirements and compatibility
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Needs: Needs Node.js
Security
What our scanner noticed· Low risk
- Scripts that your agent may run
This skill includes scripts that your agent may run. Read them before you use it.
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Published by
- Wix
- Publisher
- Wix (Official)
- Source
- wix/skills · skills/wix-headless-replatform · commit d168bc2
- Licence
- MIT License (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by Wix or Wix.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download wix-headless-replatform.zip