ClawKit Creative Studio for Lovable

ClawKit Creative Studio for Lovable is an OpenClaw plugin for turning a real website or app into launch-ready promotional material.

The core idea is simple:

Do not just scan a landing page. Let OpenClaw walk the whole product, understand the screens and flows, then create marketing assets from real evidence.

Most promotional generators start from a homepage, a URL, or a short prompt. ClawKit Creative Studio for Lovable is designed for deeper product understanding: homepage, feature pages, pricing, onboarding, dashboards, mobile views, proof pages, and approved logged-in flows.

This is an early public release. It is useful now, and it will improve continuously. Constructive feedback is welcome, especially examples where a website scan missed an important product flow or where a generated asset needed better evidence.

Creative Studio Brain

creative_studio_brain is the separate brain for this product. It is not the same as the ClawKit for Lovable brain.

• ClawKit for Lovable builds, rescues, verifies, refactors, and prepares Lovable.dev apps.
• ClawKit Creative Studio for Lovable scans the finished or preview app, understands the product journey, and creates promotional assets from evidence.

The Creative Studio Brain decides whether OpenClaw should orient the user, scan the app, audit marketability, route weak screens back to ClawKit for Lovable, create assets, review claims/privacy, or prepare a client handoff.

To match the guided experience in ClawKit for Lovable, Creative Studio also exposes a simple state-and-next-action loop:

• creative_launch_command_center is the launch cockpit: readiness score, workflow state, next action, Product Hunt button, blockers, command buttons, and launch-room export plan.
• creative_launch_room_export creates the actual launch room package: command center JSON, Product Hunt draft, social copy, gallery prompts, demo storyboard, evidence map, asset matrix, checklist, client handoff, and Lovable fix prompts.
• creative_workflow_state turns scattered launch facts into one plain status object: mode, source of truth, product status, capture status, launch risk, current blocker, and next best action.
• creative_next_action_plan chooses the next safe move so the user does not have to know whether to scan, audit, fix screens, create assets, review claims, or prepare handoff files.

Creative Studio can also create a Product Hunt launch button payload. Because Product Hunt write access is restricted and account-gated, the safe default is a one-click handoff that opens Product Hunt's submit flow with a complete launch draft for human review. If a team has approved Product Hunt write scope and server-side OAuth handling, the same payload can be used as an API-ready submission plan with a final approval gate.

Creative Studio now includes Creative Production Mode for actual media production handoff. It can decide whether a launch should stay prompt-only, render social images, render video, or produce a full media pack. The plugin still does not call image/video APIs directly; it creates renderer-ready specs for OpenClaw image-generation, design, screen-recording, FFmpeg/Remotion, or approved external video tools.

What It Does

• Asks for the app or website URL.
• Plans a full-site scan with approval gates.
• Guides OpenClaw to capture screenshots and optional video clips.
• Builds a product intelligence brief from the captured evidence.
• Scores whether the product is actually marketable yet.
• Runs Creative Audit Mode before asset generation so weak screens, missing trust signals, risky claims, and best promo angles are clear.
• Maps promotional claims back to visible evidence.
• Creates launch packs, promotional image prompts, social copy, and video storyboards.
• Creates renderer-ready social image packs, video render packs, and media asset manifests for actual PNG/JPG/MP4 production by approved tools.
• Creates agency/client-ready launch readiness reports.
• Reviews marketing claims so assets stay honest and safe.
• Includes a special Lovable Launch Pack workflow for Lovable apps.

Why It Is Different

ClawKit Creative Studio for Lovable is not a simple URL-to-video tool. It is a product-understanding workflow.

OpenClaw can browse through a complete website or app, inspect multiple pages, compare desktop and mobile views, capture evidence, and reason about user journeys. The plugin turns that ability into a repeatable creative process.

The result is a launch pack based on what the product actually shows, not generic marketing language.

Tools

Tool	Purpose
creative_studio_brain	Chooses the next Creative Studio workflow: orient, scan, audit, fix before launch, create assets, review assets, or hand off.
creative_launch_command_center	Creates the launch cockpit: readiness score, state, next action, Product Hunt button, blockers, command buttons, and launch-room export plan.
creative_launch_room_export	Creates the actual launch-room package contents for Product Hunt, social copy, gallery prompts, evidence, assets, checklist, fixes, and handoff.
creative_workflow_state	Turns messy launch facts into a simple state: mode, source of truth, product status, capture status, launch risk, blocker, and next action.
creative_next_action_plan	Chooses the next safe action: ask the user, scan, audit, fix screens, create assets, review assets, or prepare handoff.
creative_starter_guide	Explains the workflow and user options.
creative_scan_plan	Creates the full-site scan plan, evidence list, approvals, and stop conditions.
creative_browser_scan_recipe	Creates a repeatable OpenClaw browser inspection protocol.
creative_route_discovery_plan	Plans route discovery from nav, CTAs, sitemap, menus, and approved app areas.
creative_scan_session_summary	Summarizes visited routes, captures, blocked areas, errors, coverage, and next scan actions.
creative_site_intelligence	Turns page observations and screenshots into a product understanding brief.
creative_marketability_audit	Scores launch readiness and identifies weak screens, missing evidence, risky claims, and next fixes.
creative_product_audit	Runs Creative Audit Mode across product understanding, screen quality, trust signals, promo angle, and fix-before-promotion actions.
creative_evidence_map	Maps promotional claims to screenshots, page evidence, or user confirmations.
creative_capture_checklist	Creates screenshot and optional video shot lists.
creative_capture_report	Turns screenshot notes into an approved/weak/missing capture inventory.
creative_shot_selection	Selects the best screenshots for Product Hunt, social, hero, and video assets.
creative_visual_issue_report	Reports visual, mobile, copy, layout, and runtime issues that weaken launch assets.
creative_launch_asset_matrix	Maps each asset to the exact screenshot, claim, copy angle, format, and approval status.
creative_launch_pack	Produces launch copy, asset concepts, and review checklist.
creative_image_prompt_pack	Creates image-generation prompts for promotional assets.
creative_video_storyboard	Creates 15- or 30-second promo video storyboards.
creative_media_production_plan	Chooses prompt-only, rendered social image, rendered video, or full media-pack production from approved product evidence.
creative_social_image_render_pack	Creates final PNG/JPG social image render specs and prompts for Product Hunt, LinkedIn, X, hero, and campaign formats.
creative_video_render_pack	Creates MP4 video render specs: timeline, source screenshots/clips, captions, motion notes, export targets, and review checklist.
creative_media_asset_manifest	Packages media production outputs into renderer handoff files for image/video tools.
creative_export_plan	Plans the files that should make up a launch handoff pack.
creative_launch_brief	Creates a markdown launch brief from intelligence, audit, evidence, and asset matrix.
creative_prompt_export	Formats image prompts and video storyboard scenes into a handoff document.
creative_social_copy_pack	Creates LinkedIn, X/Twitter, Product Hunt, Indie Hackers, email, and banner copy.
creative_product_hunt_launch_button	Creates a guarded Product Hunt launch button payload with draft fields, submit URL, blockers, checks, and API-ready requirements.
creative_client_handoff	Creates an agency/client-ready summary with approvals and fixes.
creative_agency_report	Creates a polished client-facing launch readiness report with audit findings, fixes, asset recommendations, and launch sequence.
creative_lovable_project_context	Creates shared context from Lovable, GitHub, readiness, risks, and launch goals.
creative_lovable_to_launch_workflow	Creates the end-to-end workflow from Lovable app to launch assets.
creative_lovable_fix_prompt_pack	Converts weak launch screens and marketability issues into Lovable fix prompts.
creative_cross_plugin_handoff	Creates a handoff between ClawKit for Lovable and Creative Studio.
creative_demo_runbook	Creates a demo script that shows the product from URL to launch pack.
creative_sample_project_pack	Creates sample inputs and outputs for a fake SaaS/Lovable app.
creative_quality_scorecard	Scores a completed Creative Studio run for trust, quality, and launch readiness.
creative_user_onboarding	Creates a friendly guide with example requests, decision paths, and guardrails.
creative_lovable_readiness_feedback	Turns scan findings into Lovable prompts for screenshot-ready launch screens.
creative_lovable_launch_pack	Turns a Lovable app URL into a launch-ready marketing workflow.
creative_asset_review	Checks claims, evidence, privacy, and visual readiness before publishing.

Suggested Workflow

1. OpenClaw calls creative_launch_command_center when the user wants the full launch cockpit or Product Hunt-ready launch flow.
2. OpenClaw calls creative_studio_brain to choose the mode and next action.
3. OpenClaw calls creative_workflow_state when the facts are scattered or a session is resuming.
4. Ask the user for the website, app, preview, or Lovable URL.
5. Ask what they want: launch pack, Product Hunt gallery, social images, video storyboard, or all of the above.
6. Run creative_next_action_plan before choosing the next operational step.
7. Run creative_scan_plan.
8. Run creative_browser_scan_recipe and creative_route_discovery_plan.
9. Use OpenClaw browser tools to inspect the site and capture screenshots or approved video clips.
10. Run creative_scan_session_summary.
11. Run creative_site_intelligence with the page evidence.
12. Run creative_marketability_audit to decide whether the app is ready for promotion.
13. Run creative_product_audit to understand the product, diagnose screens, choose the promo angle, and list fixes before asset generation.
14. Run creative_evidence_map before writing strong claims.
15. Run creative_capture_checklist if more shots are needed.
16. Run creative_capture_report to inventory approved, weak, and missing shots.
17. Run creative_shot_selection to choose the best screenshots for each launch asset.
18. Run creative_visual_issue_report if screenshots reveal broken UI, poor mobile layout, weak copy, or runtime issues.
19. Run creative_launch_asset_matrix to map assets to screenshots and supported claims.
20. Run creative_launch_pack, creative_image_prompt_pack, or creative_video_storyboard.
21. Run creative_media_production_plan when the user wants actual social images or video produced.
22. Run creative_social_image_render_pack, creative_video_render_pack, and creative_media_asset_manifest for renderer-ready media production.
23. Run creative_export_plan when the user wants a reusable handoff package.
24. Run creative_launch_room_export when the user wants the complete launch-room contents.
25. Run creative_launch_brief, creative_prompt_export, creative_social_copy_pack, and creative_client_handoff as needed.
26. Run creative_agency_report when the user needs a client-ready deliverable.
27. Run creative_asset_review before publishing or sending assets to a client.

Browser Scan

Browser Scan v1 makes the full-site promise repeatable.

creative_browser_scan_recipe gives OpenClaw a desktop/mobile/app-flow inspection protocol with capture points, console checks, stop conditions, and expected outputs.

creative_route_discovery_plan tells OpenClaw where to look for meaningful routes: navigation, footer, CTAs, sitemap, feature cards, pricing buttons, app menus, and approved logged-in areas.

creative_scan_session_summary records what was actually visited, captured, blocked, or broken so the next creative tools know how complete the scan really is.

Trust Layer

The first job is not to generate pretty assets. The first job is to know whether the product is ready to market.

creative_marketability_audit checks the captured evidence and returns a readiness score, strongest screens, weak screens, supported claims, risky claims, missing evidence, recommended fixes, and asset priorities.

creative_product_audit is the full Creative Audit Mode. It combines product intelligence, scan summary, screenshots, visual issues, and marketability findings into one decision report: what the app does, who it is for, the strongest selling points, weak or broken screens, missing trust signals, best screenshots, recommended promo angle, Lovable fix prompts when relevant, and the next workflow.

creative_evidence_map connects each marketing claim to screenshots, page observations, video notes, or user confirmations. If a claim cannot be supported, the plugin marks it as risky or tells OpenClaw to remove it.

Capture Pack

Capture Pack v1 turns raw browsing evidence into a production workflow.

creative_capture_report creates a screenshot inventory with approved captures, weak captures, missing shots, and recapture recommendations.

creative_shot_selection chooses which screenshot should be used for each target asset, including Product Hunt slides, LinkedIn, X/Twitter, hero visuals, and video hooks.

creative_visual_issue_report flags visual, copy, mobile, layout, and runtime problems before they become embarrassing launch assets.

creative_launch_asset_matrix maps every target asset to the screenshot, supported claim, copy angle, format, and human approval gate.

Export Pack

Export Pack v1 turns Creative Studio output into handoff-ready content.

creative_export_plan recommends the files that should be created for a launch pack, such as launch-brief.md, asset-matrix.json, image-prompts.md, video-storyboard.md, social-copy.md, and client-handoff.md.

The plugin does not write files itself. It returns structured markdown, JSON-style data, and CSV-ready planning so OpenClaw can write files only when the user approves.

Lovable Bridge

Lovable Bridge v1 makes Creative Studio feel like the natural next step after ClawKit for Lovable.

creative_lovable_project_context gathers Lovable URL, GitHub repo, product goal, readiness summary, known stack, risks, and launch goal into a shared context.

creative_lovable_to_launch_workflow creates the exact sequence from live Lovable app to scan, fixes, rescan, launch pack, and export handoff.

creative_lovable_fix_prompt_pack turns weak screenshots, missing proof, mobile problems, and placeholder screens into focused Lovable prompts.

creative_cross_plugin_handoff makes the division of labor explicit: ClawKit for Lovable builds and stabilizes; Creative Studio scans, packages, promotes, and launches.

QA Demo Pack

QA Demo Pack v1 makes the plugin easier to trust, explain, and sell.

creative_demo_runbook creates a step-by-step demo script for showing the whole workflow from URL to launch pack.

creative_sample_project_pack creates sample input and output data so users can understand the product before scanning their own app.

creative_quality_scorecard scores a completed run across scan coverage, screenshot quality, claim support, asset matrix completeness, export readiness, and launch safety.

creative_user_onboarding explains what the plugin can do and gives users practical first requests.

Agency Report Mode

Agency Report Mode v1 turns a scan and audit into something a freelancer, consultant, or agency can send to a client.

creative_agency_report creates a polished markdown report with executive summary, product explanation, readiness score, scan evidence, strongest screens, weak or broken screens, trust gaps, recommended assets, suggested launch sequence, fix-before-publishing checklist, and optional Lovable fix prompts.

Use this after creative_product_audit, creative_capture_report, and creative_launch_asset_matrix when the user wants a commercial deliverable rather than only internal workflow output.

Lovable Extension

Lovable builds the app. ClawKit Creative Studio for Lovable launches it.

For Lovable apps, use creative_lovable_launch_pack first. It creates a workflow for scanning the Lovable preview or deployed app, capturing the best screens, identifying weak launch moments, and producing:

• Product Hunt gallery concepts.
• LinkedIn and X launch images.
• Feature cards.
• Website hero concepts.
• 15-second demo storyboard.
• 30-second walkthrough storyboard.
• Launch copy.
• Follow-up Lovable prompts if the app is not visually ready.

This pairs naturally with ClawKit for Lovable:

• ClawKit for Lovable: build, rescue, refactor, verify, ship.
• ClawKit Creative Studio for Lovable: understand, package, promote, launch.

Important Guardrails

The plugin itself does not crawl websites, read files, record screens, call image/video APIs, render MP4 files, or send network requests. It stays marketplace-safe and tells OpenClaw how to use its trusted browser, screenshot, video, image-generation, design, and renderer tools.

Do not publish generated assets until:

• Claims are visible in the product or confirmed by the user.
• No private data appears in screenshots or video.
• Broken or placeholder screens have been removed.
• The product owner has approved the message.
• Legal, brand, and platform-specific rules have been checked.

Example Requests

Scan my whole website and create a launch pack.

Turn this Lovable app into Product Hunt gallery prompts and a 30-second demo storyboard.

Review this app and tell me whether it is marketable before we create promotional assets.

Create LinkedIn, X, and website hero image prompts based on the actual app screens.

Package Shape

• src/index.ts: OpenClaw plugin entry and registered tools.
• openclaw.plugin.json: native OpenClaw manifest.
• skills/clawkit-creative-studio/SKILL.md: agent behavior policy.
• docs/tool-contracts.md: tool input/output contracts.
• docs/clawhub-listing.md: marketplace listing copy.
• docs/product-strategy.md: positioning and roadmap.