Working with evidence
Opening Brand Studio
Loading your next page.
Product / Roadmap
Launch a truthful alpha where anyone can get useful free brand evidence and confirmed users can save and continue their work.
Current alpha
Free homepage evidence. Accounts are free and save your work; the signup page says how first credits are granted during the beta. Reports cost 1 to 3 credits by depth (Snapshot, Market or Deep); a report that produces no evidence is never charged. Card checkout is not connected, and bring-your-own provider keys are not enabled on the hosted version.
Try the free readThe rebuilt paid flow uses only server-validated selected quantities for supported search, website, social and ad adapters. DataForSEO supplies the search data for paid reports and the beta weekly search check; its AI lanes stay parked, and the free read never calls it. Provider cost ceilings are separate from report credits. Paid execution runs only inside an approved maximum and the alpha limits; no automatic native-tier extras are included.
The free homepage read remains available. Accounts are free; reports run within per-report, per-account and daily limits and cost 1 to 3 credits by depth (Snapshot, Market or Deep); card checkout is not connected.
Review the alpha feedback cycle and fix any launch regressions with evidence before opening anything wider. AI and unsupported options remain excluded.
01 / Already here
A useful starting point, with the evidence behind each shipped entry.
Reads the homepage without AI and returns name, description, colors, fonts, logos, contact links and social profiles.
Implemented in /api/instant and shared with the local extractor.
Account creation, sign-in, session refresh and sign-out use Supabase Auth.
Implemented in /api/auth and the main account panel.
A signup without a Supabase session is shown as waiting for confirmation, not falsely signed in.
Implemented and covered by the existing auth contract.
Production requires confirmation and returns the email link through the Brand Studio callback.
Verified end to end with the Supabase built-in mailer and an owned Mirror Factory inbox.
Recovery email, callback, new-password form and sign-in with the changed password work as one production flow.
Verified end to end on the public alpha after correcting the GoTrue redirect parameter.
Google account choice, consent, a protected PKCE exchange and the signed-in Brand Studio handoff work as one production flow.
The web client and Supabase provider are enabled; current production proof is recorded in the canonical alpha launch receipt.
The public Vercel alpha, Instant, pricing, parking boundary, responsive pages, email flows and protected Google sign-in are verified.
Recorded in the canonical alpha launch receipt with independent auth, parking and evidence audits.
The Breakdown page explains the current paid and zero-dollar pieces by tier.
Available from the primary navigation.
Presentation note: the current Breakdown is a planning configurator. Its proposed coverage is separate from active alpha collectors and retail credits.
Adds canonical URL, structured-data types, internal and external link counts, navigation size and basic page headings at no provider cost.
Verified on the public alpha with a real Stripe pull.
Save a free read to your library, keep each brand's history and compare a new read with the last one.
Implemented in /api/brands and the brand library (hosted-app/lib/basic-brands.mjs).
Re-read a saved brand's homepage each week and say what changed: colors, logos, type, images, links and social profiles.
Implemented in hosted-app/lib/basic-watches.mjs; reads made a different way are not compared.
Download a free read as JSON or Markdown, or copy it for an assistant.
Implemented in hosted-app/lib/basic-brand-export.mjs.
Claude and ChatGPT can read saved brands and reports through a read-only connector.
Implemented in /api/mcp with OAuth sign-in; setup steps are on /mcp.
Read the pages a saved brand lists in its own sitemaps and say what changed since the last read.
Implemented in hosted-app/lib/site-pages.mjs.
Weekly social snapshots and search checks run only after the owner approves the accounts, the market and the cost ceiling.
Implemented in hosted-app/lib/social-watches.mjs and hosted-app/lib/search-watches.mjs.
The free read refuses a parked or for-sale domain instead of making a card of the parking page.
Implemented in src/local/bot-wall.mjs.
02 / In progress
Work under review. An active task is not a newly enabled collector.
Selected quantities now reach the local rankings, pages, confirmed-rival, backlink, review and site-evidence adapters. Offline tests cover boundaries, request receipts and resumability; this is not yet controlled live proof.
A renewed cost cap and confirmed target identities authorize one controlled live test; returned evidence, pending states and actual charges are reviewed before broader release.
Watch real alpha behavior for auth delivery, provider errors and report quality without enabling any parked module.
The first alpha feedback cycle is reviewed and any launch regressions have evidence-backed fixes.
brand.mirrorfactory.ai serves the current release. Whether it stays on the CT100 runtime or moves to the Vercel deployment is a separate infrastructure decision.
That decision approves and verifies the route, callback behavior and rollback plan.
03 / Planned
The next useful outcomes. These are plans, not delivery dates or current features.
Accept logos, documents, screenshots, fonts, colors, notes and social links alongside a URL.
Show what is confirmed, inferred, unknown or needs review, then ask one consequential question at a time.
Export one JSON-first package for the full report, the same shape for mature and sparse brands, with source and license inventories. Free reads already export as JSON and Markdown.
Provide controlled package pointers, versions, hashes and provenance without unrestricted folder access.
Add sourced checklists for SEO and then every report section, separating live proof from inferred behavior.
04 / Outside the current scope
These modules do not run or bill in alpha reports. A future change needs separate approval.
AI Overviews, AI visibility and LLM-mentions research are excluded while value and cost are reevaluated.
Pending value and cost reevaluation: Google AI Overviews, AI visibility and LLM-mentions research. Excluded from active reports and test budgets unless separately approved.
New images, videos and campaign concepts are not part of brand intake.
New images, video and campaign concepts are optional future tools, not intake requirements.
Turning a bitmap logo into SVG is an optional future production step.
PNG or JPG logo conversion to SVG is an optional improvement after the original asset is preserved.
Background removal, repair, upscaling and reconstruction are outside this alpha.
Background removal, artifact repair, upscaling and asset reconstruction remain optional production steps.
Additional proprietary research needs its own approval and price preview.
Context.dev, automatic keyword brainstorming and other unsupported research extras remain planning-only. They require separate implementation, pricing and approval and are not bundled into the selected-plan flow.
Roadmap record updated 2026-10-22. This page renders content/roadmap.json; source records describe their original verification, not a new test run.