How well does it score?

89% average fidelity over 4 recorded runs. Best model: GLM-4.6 (CN). Weakest recorded run: 88% on Codex.

Are failures included?

No build failures are recorded for this prompt, and the log says so rather than implying every model succeeded elsewhere.

Concept render — composed live from this prompt's palette & layout. Reproducible with the library, not a stock screenshot.

Concept scene: Liquid Glass — open in library →

HiringVerifiedby Motif Studiopublished 2026-09-10

Recruiting studio — open roles with culture evidence

Q: What does the Recruiting studio — open roles with culture evidence prompt produce?

A hiring website with Open roles, Culture evidence, Referral promise, Process page. The vibe target is "evidence-first, human, referral-true" and the prompt body is printed in full on the page.

vibe: evidence-first, human, referral-true

starter — 4 blocks, 2 stacks≈ 700 tokens to run (estimate)

prompt.txt — copy-paste ready

A recruiting studio's careers page that proves culture with evidence instead of a values poster.
• Open roles: each role as a card with the team, the level, and an honest salary band — 'we tell you the band before you interview' as the opening trust move.
• Culture evidence: three pieces of proof rather than adjectives — a real project the team shipped, a real retro note that changed how they work, and a real day-in-the-life photo with captions.
• Referral promise: 'refer someone we hire and we pay you two months later, in cash, no clawback' — the promise stated with its actual terms.
• Process page: the interview journey in four numbered steps with who you meet at each and how long each wait is; ghosting is named as the industry failure this studio refuses.
• The no-mansplaining clause: hiring manager notes on what the role will not ask you to do.
• Palette: ink, paper, one signal colour reserved for apply buttons.

Stacks shipped:ReactHTMLBlocks:Open rolesCulture evidenceReferral promiseProcess page

Community:▶ Re-run this prompt ⤴ Submit a remix of it ★ Your saved listRe-runs are a labelled simulation; the run log on this page is the real record.

⚠ Before you run it, know this