CyberSmart

Cinevva's analyzer read this game's design doc and its source, then wrote the brief below. "Make your own" hands it to the game creator as your opening message, so you start from a real spec rather than a one-line idea.

Rebuild a responsive browser-based educational decision game titled “CYBERSMART: THINK BEFORE YOU CLICK,” presented as a dark neon cyber-safety mission dashboard rather than an action game. The game is a playable set of scenario questions: players investigate fictional online-safety situations, choose a response, and submit a raw call without instant right/wrong scoring. Include ten open mission zones plus a final mission called “THE PHISHER,” for 11 total missions.

Visual design and layout: use a deep navy/black background with subtle radial gradients, translucent indigo glass panels, thin blue-violet borders, rounded corners, soft shadows, and bright cyan, purple, green, pink, and yellow accents. The top bar contains a rounded gradient square “CS” logo, the title “CYBERSMART: THINK BEFORE YOU CLICK,” a small “Cyber Detectives · Live Mission Game” subtitle, a room-code input, Join button, offline/room status, and two mode buttons: “Player View” and “Host Panel.” On desktop, use a two-column layout with a left detective sidebar and a scrollable main content area; on narrow screens hide the sidebar and stack the content. The sidebar shows a “Detective profile” card for Alex with an A avatar, “nickname · locked in,” followed by a “Mission Map” list with numbered mission buttons and a star for the final mission. Completed missions show a green completion treatment.

The main player view begins with a large hero card: eyebrow text “Presented by Heshna Nachiyar Suresh · 9J,” headline “Pause. Investigate. Then act.,” and explanatory copy saying there are ten open missions plus The Phisher, with no gates or gotcha scoring during play; the player submits a raw call and keeps moving. On the right show a large progress counter such as “0/11” and “missions touched.” Below it, show a mission map progress card with the current room (or OFFLINE), a numeric touched count, and a horizontal gradient progress bar. Display mission cards in a responsive grid. Each card has its zone label, title, short description, completion status, and an “Open mission” button; completed cards change to “Review call.” Give the final mission a pink-accented special card.

The ten mission scenarios and their exact decision content are:
1. “Phishing & Suspicious Links”: a fictional SecureAlert says the account will be locked in 10 minutes, includes a surprise link, and asks the user to sign in. Choices: “Open link” / “Fastest way to fix it”; “Reply” / “Ask the sender what is going on”; “Stop & verify” / “Use the official app/site or trusted adult.” The safe answer is Stop & verify.
2. “Online Scams & Social Engineering”: compare messages and identify risky cues. Choices: “School update only” / “A normal update in the usual channel”; “Account threat + prize offer” / “Fear, urgency, impersonation, and a too-good-to-be-true reward”; “All messages are safe” / “A familiar logo means zero risk.” The safe answer is the account threat and prize offer.
3. “Password Security”: Rohan reuses one password for gaming, email, and cloud storage. Choices: “Reuse it everywhere” / “Easy to remember”; “Avoid reuse, no plan” / “I will just try to remember later”; “Unique passwords + habits” / “Use a password manager and turn on MFA.” The safe answer is unique passwords plus habits.
4. “Multi-Factor Authentication (MFA)”: a verification code arrives when the player did not try to log in. Choices: “Approve it” / “Someone must be helping”; “Share the code” / “They said they are support”; “Deny, change password, verify” / “Treat it as a warning sign.” The safe answer is deny, change password, and verify.
5. “Social Media Privacy”: Maya posts a photo showing her school badge, a landmark, and that she is out right now. Choices: “Nothing — it is just a photo” / “Images cannot reveal details”; “It reveals routine, place, and identity clues” / “Small details can combine”; “Only the caption matters” / “Background details do not count.” The safe answer is that it reveals routine, place, and identity clues.
6. “Digital Footprint”: connect an ordinary post, comment, photo, like, and profile. Choices: “Post first, think later” / “Delete it if needed”; “Pause and consider the bigger picture” / “Would I be okay with this later?”; “Only worry about public posts” / “Private content never travels.” The safe answer is pause and consider the bigger picture.
7. “AI Privacy & Responsible AI Use”: trim Aarav’s over-detailed prompt and fact-check output. Choices: “Include name, school, address, homework photo” / “More detail always means better AI”; “Remove personal details and classmates’ work” / “Use a generic description and your own work”; “Trust the first answer” / “AI sounds confident, so it is correct.” The safe answer is remove personal details and classmates’ work.
8. “Fake Messages & Impersonation”: a friend’s account urgently asks for a gift card code. Choices: “Reply in the same chat” / “The name looks familiar”; “Use another trusted channel to confirm” / “Call or message through a known route”; “Send a small amount first” / “Test it with less risk.” The safe answer is use another trusted channel.
9. “Safe Browsing & Verification”: a familiar-looking page unexpectedly asks for a password and full address. Choices: “Fill it in fast” / “The page looks polished”; “Close and verify through the official source” / “Use a bookmark or type the official address”; “Partially fill it in” / “Half the details is half the risk.” The safe answer is close and verify through an official source.
10. “STOP · CHECK · VERIFY · THINK · ACT”: a fresh fictional message says, “Your teammate needs your login now — deadline in 2 minutes.” Choices: “Act immediately” / “They sound urgent”; “Stop, check, verify, think, then act” / “Slow down and confirm independently”; “Ignore every message forever” / “Safety means never responding.” The safe answer is the five-step method.
Final “THE PHISHER”: a five-round timed-gauntlet concept mixing all mission topics, with the visible first round asking about the suspicious URL “example.invalid/securealert” requesting a login. Its visible choices are “Click — it says SecureAlert” / “The logo looks official”; “Stop & verify through the official source” / “Do not use the suspicious link”; “Reply with my password” / “Support asked.” The safe visible answer is stop and verify.

When a player opens a mission, show a panel with the zone label plus “raw answer mode,” title, scenario prompt, a Map back button, and tags reading “No instant right/wrong,” the room/offline status, and “Change before reveal.” Present the choices as large selectable cards labeled A, B, and C. Selecting one highlights it and immediately displays a green confirmation strip reading “Locked in ✅ Your answer is syncing to the host room.” Keep the answer changeable before the host reveal; do not show correctness in player mode. Include a “Progress only: X/11 touched” footer and a “Back to map” button. Do not gate later zones: all zones are unlocked from the start.

Implement the room interaction shown by the interface: accept a 4–6 character uppercase room code, persist it locally, show the current room or OFFLINE, and synchronize players’ nicknames/avatars and raw answers for the same room using browser-local persistence and a same-room broadcast mechanism. The default player is Alex with an A gradient avatar. Player progress is the number of missions answered, not a score.

The Host Panel mode replaces the player content with a live receipts dashboard. Show “Host Panel · room [code],” a heading “Live receipts, no scores — yet,” text telling the host to share the room code, and a prominent “END MISSIONS” button (which changes to “MISSIONS ENDED”). Show joined player cards with avatar, name, and touched count. Below, show a horizontally scrollable matrix table: each row is a detective, each mission is a column, and cells show “not touched” or the selected answer letter and choice text. Players in the same room should appear as synced players. When the host ends missions, unlock a reveal section titled “Reveal unlocked 🔓,” explaining that correct answers and safe-action receipts are visible. For each mission, show the host’s call, mark it “Correct” in green or “Needs a rethink” in pink, and display the relevant safety explanation. The explanations should emphasize urgency/unexpected links as phishing, independent verification, unique passwords and MFA, denying unexpected codes, privacy clues in photos, pausing before posting, removing personal data from AI prompts and checking facts, using a separate trusted channel, closing unexpected forms, and the STOP/CHECK/VERIFY/THINK/ACT method. Keep the overall tone supportive and educational, with no competitive scoreboard during play.