Skip to content

Finish remaining council marketplace follow-up items - #115

Merged
profilesearch merged 5 commits into
mainfrom
fix/finish-council-simplification-items
Jul 22, 2026
Merged

Finish remaining council marketplace follow-up items#115
profilesearch merged 5 commits into
mainfrom
fix/finish-council-simplification-items

Conversation

@profilesearch

@profilesearch profilesearch commented Jul 22, 2026

Copy link
Copy Markdown
Owner

Objective

Close the remaining council recommendations from the marketplace-first public-site simplification without mutating applications, jobs, orders, payments, refunds, payouts, or settlement state.

Changes

  • Reserve executive analytics events (generate_lead, qualify_lead, close_convert_lead, purchase) for successful persisted outcomes rather than intent-only static CTA clicks.
  • Prevent post_task_cta_click from being attached to links that do not open a task draft.
  • Qualify FAQ checkout, Stripe, publication, and privacy-sharing language.
  • Align Pricing and Starter Offers around four canonical offers and synchronize draft budgets:
    • AI Output Verification — $99
    • Automation QA Sprint — $199
    • Clay/GTM QA Sprint — $199
    • Real-World Check — $79
  • Replace unsourced competitor-rate claims on Pricing with a generic, verify-current-terms comparison.
  • Render service provider, worker rating/review count, and delivery facts from canonical API fields; omit unknown provider/delivery claims and distinguish unknown review facts from known zero-review listings.
  • Add regression coverage for all of the above.

Independent review resolution

  • Unknown (NULL) review facts render as unavailable; known zero reviews remain eligible for the “New listing” treatment across browse, detail, homepage preview, and seller My Services surfaces, with direct browser coverage of each state.
  • Every explicit service delivery promise must be an integer from 1–365. An omitted/null fixed-service delivery value is canonicalized to the bounded seven-day legacy default, preserving direct-API compatibility; historical fixed rows with NULL retain the same checkout fallback. Legacy zero/invalid values fail closed in public UI rather than becoming same-day or “Flexible.”
  • Service pricing updates validate the fully merged persisted/request state: fixed requires a positive fixed price, hourly requires a positive hourly rate, rejected transitions do not mutate the listing, and valid fixed/hourly/custom transitions clear stale amount-family fields and remain checkout-compatible.
  • Service updates preserve the same AI/API endpoint invariant as creation, and the seller form marks the endpoint required for API-fulfilled AI services.
  • Unknown provider types use neutral UI copy and no Human/Agent badge; explicit known types retain their labels.
  • Pricing routes all four canonical offers directly to the exact final task draft, with browser checks of final title/budget; website-QA and lead-research SEO pages retain separate semantically correct drafts.
  • Pricing JSON-LD qualifies checkout-dependent fee claims just like visible copy.
  • Public suggested ranges include their canonical defaults.
  • FAQ sharing language matches the current Privacy Policy rather than attributing broader sharing to it.
  • Public intent/template CTAs are guarded from executive conversion events, including SPA inline CTA markup.
  • Root Playwright output directories are ignored so failed-run artifacts cannot be accidentally packaged.

Verification

  • Backend: 396 tests passed.
  • Browser: 108 passed, 6 skipped across desktop/mobile.
  • Local release verifier: 20/20 route/viewport checks plus exact FAQ, pricing, budget, direct offer-funnel, SEO use-case, null-review, legacy-delivery, provider, and service-contract assertions.
  • Security static checks: passed.
  • Public shell assets: passed.
  • Public shell synchronization: 0 static page(s) pending.
  • Performance budgets: passed.
  • npm audit: 0 vulnerabilities.
  • git diff --check: passed.
  • Added-diff secret scan: no private-key/provider-secret patterns; four generic token= matches were manually confirmed as synthetic test-auth arguments.

Safety boundaries

  • The patch changes future service-listing validation, but validation/review did not mutate production marketplace or payment state.
  • No applications are accepted or funded.
  • No jobs, orders, payments, refunds, payouts, or settlement state are created or changed.
  • No external communications or social posts are sent.

@vercel

vercel Bot commented Jul 22, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
gohire-humans Ready Ready Preview, Comment Jul 22, 2026 7:33am

Request Review

@profilesearch
profilesearch merged commit 5d9386c into main Jul 22, 2026
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant