The public funnel now does a better job showing what ApplyBench actually does before asking a cold visitor to sign up.

• The landing demo can now layer a limited pasted-resume preview on top of the public job-risk check for the same role.
• Public job-check results now change the next step based on the verdict: tailor, verify first, or skip instead of using one generic call to action.
• Refreshed the public builder-notes page with clearer architecture, trust-boundary, and preview-versus-production diagrams.
• Tightened landing copy so the public path reads more clearly as check the role first, then continue into Resume Analyzer when it looks worth pursuing.

The signed-in job inbox now helps users paste links, start from curated public sources, and see clearer source health before spending credits.

• Expanded starter sources across AI labs, voice AI, cloud/platform, developer tools, enterprise SaaS, automation, fintech, health, and remote-first categories.
• Improved saved-search summaries so users can see saved pages, roles found, saved roles, scan issues, last scan status, and the last useful result.
• Made scanner failures clearer by showing which source failed, why it failed, and what to try next.
• Improved closed-posting detection for imported job links, including LinkedIn listings that say they are no longer accepting applications.
• Updated active application cards so each role leads with one recommended next action before showing other tools.
• Clarified that saved searches are user-controlled manual scans, not broad job-board crawling, scheduled alerts, or auto-apply.
• Added an internal human test guide for the Opportunity Command Center and browser-capture extension.

ApplyBench now has the first slice of a role-centered command center for ranking jobs before users spend application effort.

• Added a signed-in ranked job inbox for checking up to 10 public job links at once.
• Added Opportunity Fit verdicts that combine trust, timing, fit, eligibility, and strategy hints.
• Added a saved Candidate Strategy Profile first version for target roles plus preferred and avoid signals.
• Added dashboard health and story-bank starter previews from existing analyzer data.
• Added scanner support for user-provided Greenhouse, Lever, Ashby, SmartRecruiters, Workable, and visible company careers pages.
• Added grouped ranking explanations and save-to-active-queue for promising roles.
• Added a first trusted role-discovery loop for manually revisiting saved public sources.

ApplyBench is live in soft launch with a stronger job-trust entry point and clearer application workspace.

• Enabled the public job scam and ghost-job check as the first low-friction action.
• Added clearer wait-state copy for long-running resume and application tools.
• Improved active queue and archive layouts so application cards stay visually even.
• Added a plain ATS-safe resume export preview so downloaded PDFs match expectations.

Follow-on tools now keep the role context from the original analysis instead of feeling like separate islands.

• Dashboard actions preserve the originating role when opening ATS Checker, Cover Letter, Interview Prep, Follow-Up, or Salary Negotiation.
• Resume Analyzer results lead with a clearer Should I apply? decision and Top 3 fixes.
• Work authorization language was tightened to stay grounded in posting evidence, not legal advice.

The product narrowed around checking job quality first, then improving the application only when it is worth the effort.

• Added a public trust-only job check path.
• Shifted pricing copy toward pay-per-application decisions instead of subscription framing.
• Soft-depromoted legacy tools from the main public workflow while keeping existing saved results available.