Eight complete mechanism cards
Minimum tests and stop conditions below are proposals, not executed experiments. No game was launched. Team size is unknown; an Indie tag is not team evidence.
We Were Here Tomorrow
Steam · Total Mayhem Games · 2026-10-20T23:12:59.155Z
Store date 2026-10-20 (timezone unspecified); purchasable with an Out Now announcement. US regular price $19.99, current $17.99. CNY regular price unknown. All-language reviews: Mixed, 114.
- Input and feedback
- Official description: two players have different information and tools, describe clues over walkie-talkies and operate the facility together. Progress opens obstacles and areas. Not playtested.
- Goals
- Solve time, gravity and space puzzles and investigate the facility. Co-op only: two players, working microphones and an internet connection are required.
- Progress and resource economy
- The description establishes level and narrative progression. Currency flows, complete tool unlocks and failure penalties are unknown; store copy does not establish retention.
- Ads and natural breaks
- US regular price $19.99; launch price $17.99 until the displayed November 4. CNY regular price is unknown. In-game ads were not verified. A solved puzzle is a hypothetical natural break for an original short-level design.
- H5 production constraints, hypothesis
- Full 3D co-op, networking, voice and complex environments are not a direct low-cost port. Independently test one original fixed-camera puzzle with different clues on two screens, avoiding camera rotation.
- Player need evidence
- Three visible reviews in Recent order and the current language included two substantive comments. A recommender found some early puzzles clear; a non-recommender found some sequences drawn out. This is not a representative sample of all 114 ratings.
- Pain evidence
- Both substantive comments reported progression blocks requiring restarts and motion sickness; these are player reports, not reproduced bugs. The recommender also described possible sequence skipping with the new gun system. Complaints are coded separately from recommendation labels.
- Minimum validation hypothesis
- Give two players separate symbol rules and a control panel. Test one unlock without shared screens; record misunderstandings, waiting and whether they can explain the result.
- Stop condition
- Reject the information-asymmetry prototype if one player can solve it alone or connection/waiting costs overwhelm collaboration. Launch chart position is not validation.
- Applicability
- A structural reference for communication puzzles, not a verified low-price primary recommendation or confirmed H 5 opportunity.
Normal-sample original reviews: 1 · 2
Back to brief focus index · Back to rank focus index
Team size and code/asset licences are unknown. Public descriptions were studied without playtesting. Implement mechanics independently and use owned or licensed assets.
桌面破坏神：挂机ARPG / Deskrawl
Steam · First Day Games · 2026-10-20T23:20:30.245Z
Store date 2026-09-28 (timezone unspecified); released. US regular price $6.99, current $6.99. CNY regular price unknown. Traditional Chinese reviews: Mixed, 364.
- Input and feedback
- Four classes auto-fight and collect loot; captured enemies and bosses become followers that collect items, buff or support combat.
- Goals
- Use the officially described loop or match goal; failure rules and the complete experience were not playtested.
- Progress and resource economy
- Gold and materials strengthen equipment; experience levels characters; investing in three towns unlocks services. The official description also includes Steam-market trading, which is not assumed portable to H 5.
- Ads and natural breaks
- In-game ads, frequency caps and paid economy were not tested. Stage or match completion is only a hypothetical natural break for an original design. CNY regular price is unknown; the approximately RMB 60 preference is not confirmed.
- H5 production constraints, hypothesis
- Test only choose build → one automated encounter → compare loot utility. Do not promise real-money returns; background operation, trading and large item sets expand cost.
- Player need evidence
- The first 10 raw cards contain 9 substantive comments; the brief “already earned it back” claim does not establish mechanics. Substantive comments praise builds and balancing while criticizing scarce drops and thin endgame; sales/profit claims are unverified player statements.
- Pain evidence
- The first 10 raw cards contain 9 substantive comments; the brief “already earned it back” claim does not establish mechanics. Substantive comments praise builds and balancing while criticizing scarce drops and thin endgame; sales/profit claims are unverified player statements.
- Minimum validation hypothesis
- Test only choose build → one automated encounter → compare loot utility. Do not promise real-money returns; background operation, trading and large item sets expand cost. These are research proposals, not tests executed this issue.
- Stop condition
- Reject if automation offers no meaningful decisions or rewards depend only on extremely rare drops.
- Applicability
- Study public mechanics and feasibility of independent implementation; do not copy code, art, audio or story.
Normal-sample original reviews: 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9
Back to brief focus index · Back to rank focus index
Team size and code/asset licences are unknown. Public descriptions were studied without playtesting. Implement mechanics independently and use owned or licensed assets.
Scam With Your Friends
Steam · JakeHub, Metater · 2026-10-20T23:22:20.618Z
Store date 2026-10-16 (timezone unspecified); Early Access. US regular price $9.99, current $8.99. CNY regular price unknown. All-language reviews: Mostly Positive, 2662.
- Input and feedback
- Official fictional comedy: real-time AI dialogue, co-op, workplace interruptions and daily quotas. This research covers interaction structure, not real-world scam methods.
- Goals
- Use the officially described loop or match goal; failure rules and the complete experience were not playtested.
- Progress and resource economy
- Run progression adds workday scenarios and interruptions; AI dialogue requires an external service and internet. Actual inference cost, latency and long-term economy are unknown.
- Ads and natural breaks
- In-game ads, frequency caps and paid economy were not tested. Stage or match completion is only a hypothetical natural break for an original design. CNY regular price is unknown; the approximately RMB 60 preference is not confirmed.
- H5 production constraints, hypothesis
- For original customer-service or negotiation comedy, test bounded text choices and fictional characters before voice. Do not collect real personal data, clone real voices or reproduce scam scripts.
- Player need evidence
- All 10 Recent reviews were read. Both recommenders and non-recommenders report Chinese speech recognition issues. Some enjoy conversations and recap recordings; others report service failures, crashes, white screens and request saves. Regional restrictions are player claims, not independently verified.
- Pain evidence
- All 10 Recent reviews were read. Both recommenders and non-recommenders report Chinese speech recognition issues. Some enjoy conversations and recap recordings; others report service failures, crashes, white screens and request saves. Regional restrictions are player claims, not independently verified.
- Minimum validation hypothesis
- For original customer-service or negotiation comedy, test bounded text choices and fictional characters before voice. Do not collect real personal data, clone real voices or reproduce scam scripts. These are research proposals, not tests executed this issue.
- Stop condition
- Stop if service failure lacks a fallback, recognition errors punish players or safety boundaries cannot be maintained.
- Applicability
- Study public mechanics and feasibility of independent implementation; do not copy code, art, audio or story.
Normal-sample original reviews: 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10
Back to brief focus index · Back to rank focus index
Team size and code/asset licences are unknown. Public descriptions were studied without playtesting. Implement mechanics independently and use owned or licensed assets.
Moon Sweepers
Steam · DragDrop Games · 2026-10-20T23:23:56.825Z
Store date 2026-10-23 (timezone unspecified); planned release, unreleased. No public price displayed. CNY regular price unknown. All-language reviews: No user reviews, 0.
- Input and feedback
- Start with a toothbrush and visibly clean lunar dust, then haul full loads to a collection station and launch them for payment. Up to 16-player co-op is an announced feature.
- Goals
- Use the officially described loop or match goal; failure rules and the complete experience were not playtested.
- Progress and resource economy
- Sell dust to buy tools, expand carrying capacity, place collection stations and unlock vehicles; cleaned area provides visible progress.
- Ads and natural breaks
- In-game ads, frequency caps and paid economy were not tested. Stage or match completion is only a hypothetical natural break for an original design. CNY regular price is unknown; the approximately RMB 60 preference is not confirmed.
- H5 production constraints, hypothesis
- An original H 5 hypothesis uses one small solo area: clean → fill capacity → deliver → one upgrade. Test distance-versus-capacity decisions before promising networking.
- Player need evidence
- Unreleased with no user reviews; 16-player co-op and the feel of feedback were not tested.
- Pain evidence
- No current normal-user comments substantiate pain points. Unreleased status or low attention is not failure.
- Minimum validation hypothesis
- An original H 5 hypothesis uses one small solo area: clean → fill capacity → deliver → one upgrade. Test distance-versus-capacity decisions before promising networking. These are research proposals, not tests executed this issue.
- Stop condition
- Reject if before/after cleaning is unclear or upgrades merely extend decision-free clicking.
- Applicability
- Study public mechanics and feasibility of independent implementation; do not copy code, art, audio or story.
Back to brief focus index · Back to rank focus index
Team size and code/asset licences are unknown. Public descriptions were studied without playtesting. Implement mechanics independently and use owned or licensed assets.
Groundwork Simulator
Steam · XYZ STUDIO · 2026-10-20T23:24:29.536Z
Store date 2026-10-22 (timezone unspecified); planned Early Access, unreleased. No public price displayed. CNY regular price unknown. All-language reviews: No user reviews, 0.
- Input and feedback
- Excavate trenches, pile soil and haul material. Drill for deposits, then use sluices and a shaking table; reshaped ground affects later routes.
- Goals
- Use the officially described loop or match goal; failure rules and the complete experience were not playtested.
- Progress and resource economy
- Pumps and hoses supply water; trucks transport machinery; hydraulic resistance and muddy traction shape control. Gold prices and upgrade curves are unknown.
- Ads and natural breaks
- In-game ads, frequency caps and paid economy were not tested. Stage or match completion is only a hypothetical natural break for an original design. CNY regular price is unknown; the approximately RMB 60 preference is not confirmed.
- H5 production constraints, hypothesis
- Test discrete dig, transport and sort steps, not continuous terrain or hydraulic simulation; first test feedback on deposit selection.
- Player need evidence
- Unreleased with no user reviews. The developer explicitly warns that soil/water physics are CPU-heavy; appearance is not evidence of low production cost.
- Pain evidence
- No current normal-user comments substantiate pain points. Unreleased status or low attention is not failure.
- Minimum validation hypothesis
- Test discrete dig, transport and sort steps, not continuous terrain or hydraulic simulation; first test feedback on deposit selection. These are research proposals, not tests executed this issue.
- Stop condition
- Stop the H 5 direction if its appeal disappears without high-fidelity physics.
- Applicability
- Study public mechanics and feasibility of independent implementation; do not copy code, art, audio or story.
Back to brief focus index · Back to rank focus index
Team size and code/asset licences are unknown. Public descriptions were studied without playtesting. Implement mechanics independently and use owned or licensed assets.
泡泡镇餐厅物语 / PAO PAO
Steam · ArtDock · 2026-10-20T23:24:49.853Z
Store date 2026-10-26 (timezone unspecified); planned Early Access, unreleased. No public price displayed. CNY regular price unknown. All-language reviews: No user reviews, 0.
- Input and feedback
- Cook, serve guests and arrange a restaurant solo or with up to four players; farm, fish, gather and interact with NPCs outside.
- Goals
- Use the officially described loop or match goal; failure rules and the complete experience were not playtested.
- Progress and resource economy
- Restaurant operations and gathering drive progression with character/location customization. Deeper progression, areas and seasonal activities are future plans, not verified released content.
- Ads and natural breaks
- In-game ads, frequency caps and paid economy were not tested. Stage or match completion is only a hypothetical natural break for an original design. CNY regular price is unknown; the approximately RMB 60 preference is not confirmed.
- H5 production constraints, hypothesis
- Limit a test to one menu, two recipes and one restocking cycle; multiplayer, an open town and farming/fishing are not a minimum H 5 scope.
- Player need evidence
- The main game is unreleased with no user reviews. A demo entry was visible but was not downloaded or played; feel remains unverified.
- Pain evidence
- No current normal-user comments substantiate pain points. Unreleased status or low attention is not failure.
- Minimum validation hypothesis
- Limit a test to one menu, two recipes and one restocking cycle; multiplayer, an open town and farming/fishing are not a minimum H 5 scope. These are research proposals, not tests executed this issue.
- Stop condition
- Stop if appeal requires stacking many systems or serving offers no clear efficiency trade-off.
- Applicability
- Study public mechanics and feasibility of independent implementation; do not copy code, art, audio or story.
Back to brief focus index · Back to rank focus index
Team size and code/asset licences are unknown. Public descriptions were studied without playtesting. Implement mechanics independently and use owned or licensed assets.
Tidy Up the Halloween Shop
Steam · Over The Moon Games · 2026-10-20T23:25:12.487Z
Store date 2026-10-24 (timezone unspecified); planned release, unreleased. No public price displayed. CNY regular price unknown. All-language reviews: No user reviews, 0.
- Input and feedback
- Pick up pumpkins, sweets and potions and return them to shelves, carrying multiple items; follow odd clues to a mysterious friend, without a timer.
- Goals
- Use the officially described loop or match goal; failure rules and the complete experience were not playtested.
- Progress and resource economy
- Restored order and a mystery reveal are the stated rewards. Scores, currency and upgrade systems were not described and are not invented.
- Ads and natural breaks
- In-game ads, frequency caps and paid economy were not tested. Stage or match completion is only a hypothetical natural break for an original design. CNY regular price is unknown; the approximately RMB 60 preference is not confirmed.
- H5 production constraints, hypothesis
- Use owned art for one shelf, a few objects and an explainable clue. Promotional footage must show real placement, not imply absent horror mechanics.
- Player need evidence
- Unreleased with no user reviews. The developer discloses generative AI across several disciplines with human editing; this does not grant asset-copying rights.
- Pain evidence
- No current normal-user comments substantiate pain points. Unreleased status or low attention is not failure.
- Minimum validation hypothesis
- Use owned art for one shelf, a few objects and an explainable clue. Promotional footage must show real placement, not imply absent horror mechanics. These are research proposals, not tests executed this issue.
- Stop condition
- Stop expansion if placement cannot be inferred from clues or only the seasonal appearance, not tidying feedback, attracts players.
- Applicability
- Study public mechanics and feasibility of independent implementation; do not copy code, art, audio or story.
Back to brief focus index · Back to rank focus index
Team size and code/asset licences are unknown. Public descriptions were studied without playtesting. Implement mechanics independently and use owned or licensed assets.
Figure Me Out
Steam · Solo Groove Games · 2026-10-20T23:25:36.780Z
Store date 2026-11-03 (timezone unspecified); planned release, unreleased. No public price displayed. CNY regular price unknown. All-language reviews: No user reviews, 0.
- Input and feedback
- Read an opponent’s forehead number while moving and type it. Smoke, flashes and distractions obstruct reading; binoculars and temporary photos aid memory. This is not shape merging or a title-inferred puzzle game.
- Goals
- Use the officially described loop or match goal; failure rules and the complete experience were not playtested.
- Progress and resource economy
- Deathmatch, team deathmatch and capture the flag support up to 20 players. Sightlines, memory and gadgets define the stated interaction; long-term economy is unknown.
- Ads and natural breaks
- In-game ads, frequency caps and paid economy were not tested. Stage or match completion is only a hypothetical natural break for an original design. CNY regular price is unknown; the approximately RMB 60 preference is not confirmed.
- H5 production constraints, hypothesis
- First test an original single-screen reveal → occlusion → type interaction. Multiplayer requires touch-input fairness checks; do not promise a 20-player FPS immediately.
- Player need evidence
- Unreleased with no user reviews; input fairness, latency and mobile-keyboard obstruction were not tested.
- Pain evidence
- No current normal-user comments substantiate pain points. Unreleased status or low attention is not failure.
- Minimum validation hypothesis
- First test an original single-screen reveal → occlusion → type interaction. Multiplayer requires touch-input fairness checks; do not promise a 20-player FPS immediately. These are research proposals, not tests executed this issue.
- Stop condition
- Reject mobile H 5 if typing overwhelms observation or legibility depends on a large high-resolution screen.
- Applicability
- Study public mechanics and feasibility of independent implementation; do not copy code, art, audio or story.
Back to brief focus index · Back to rank focus index
Team size and code/asset licences are unknown. Public descriptions were studied without playtesting. Implement mechanics independently and use owned or licensed assets.
Ten observations after all-candidate screening
All 384 upstream display records were screened for themes, actions, windows and cooling. No theme was guessed for 84 unverified titles. Ten priority families are not ten independent hotspots. Visibility alone does not establish reachable demand or a supply gap, so deep research remains 0 and no project is proposed.
Theme observation · new_in_observed_prefix
Relationship-probability fiction newly appears near the front of Bilibili
Visible at position 3 with 827,000 displayed plays, by Mr-Ferret. The title turns an invisible social outcome into a number. Full-video context and normal comments were not reviewed; this is not evidence of demand for mind-reading gameplay or real relationship prediction.
Hypothesis only: observable fictional clues change a relationship meter under explicit rules. Choices need causal feedback; never promise predictions about real people.
Official source
我得到了，能看到表白率的超能力……
Trace between rank changes and full research
Theme observation · new_in_observed_prefix
A lost-property verification puzzle appears at Weibo No. 1
The Weibo title describes a finder leaving a question for the owner of RMB 38,000, with displayed heat 648,613. Only the chart title was verified; the question, solution and event details were not. A trending title is not independently verified news or a game-supply gap.
Hypothesis: verify ownership of fictional lost property using non-sensitive details, with evidence-consistency feedback. Reject a one-puzzle test if success depends on guessing the author rather than reasoning.
Official source
男子捡到38000元留了道题给失主
Trace between rank changes and full research
Theme observation · new_and_continuing
Queues are visible in both news search and restaurant commentary
A related Baidu title is newly observed at No.14 with index 6,568,700. The Bilibili restaurant video remains at position 5 while plays move 478,000→894,000. These share a queue setting, not a verified event or demand for a queue-management game; metrics are not added.
Hypothesis: make ordering, serving and waiting tolerance visible trade-offs; footage should show one queue being resolved. Verify video context before attributing motives to a real owner.
Official source
最害怕顾客排队的老板出现了 · 特厨隋坡|排队超久的网红店？！今天我们来说实话！！
Trace between rank changes and full research
Theme observation · new_in_observed_prefix
Waste-disposal embarrassment appears at Baidu No.8
The title about embarrassment when disposing of rubbish is at No.8 with index 7,135,357. The label is visible, but causes and audience size are unverified; it is not a mental-health diagnosis or proof of widespread demand.
Hypothesis: tidy and sort objects in a fictional shared home while protecting personal belongings; reward a clean space and clear completion. Stop expansion if clicks reflect voyeurism rather than interest in the actions.
Official source
00后北漂患上“扔垃圾羞耻”
Trace between rank changes and full research
Theme observation · creator_family_observed
Two creators show a shared celebrity-remix family
The Gentleman remix moves 41→2 and 65,000→344,000 plays; Uptown Funk moves 4→25 and 1.554m→1.810m. Different creators still share one celebrity-remix family, not two independent new memes. Falling position alongside rising plays illustrates the distinction.
Keep only the abstract structure of rhythmic mismatch and contrast. Do not reuse the celebrity likeness, songs or clips. Without licensed assets and a distinct playable promise, do not enter production validation.
Official source
⚡赵本山 《绅士Gentle Man》⚡ · 赵本山携手象牙山巨头们激情演唱《Uptown Funk》
Trace between rank changes and full research
Theme observation · predictable_event_window
First snow is a seasonal window, not new gameplay demand
Douyin moves 10→11 while heat changes 9.291m→9.399m. The title supports seasonal-scene visibility. Local weather, the next snowfall and future reach were not independently checked, so no definite 3–7-day forecast is made.
Hypothesis: clear snow, open paths and reveal architectural detail in an original small scene. Footage must show the actual before-and-after action. Seasonal clicks do not establish continued play.
Official source
当初雪落在大同古城
Trace between rank changes and full research
Theme observation · new_coverage_not_new_entry
Dorm survival and health-bar fiction are visible in suspense fantasy
The top three of female reading · suspense fantasy displayed 在读 values are 30.5万, 25.9万 and 25.7万 (the platform’s rounded reading metric, not independent readers), cut off at October 20 24:00. This category is newly covered, not evidence of new entries. “Watching ads to grow stronger” is fiction, not game ad acceptance; reading is not game users.
Extract only hidden information, limited resources and evasion. A possible test starts with an original single-room level without forced ads. Novel titles do not authorize copying characters or plots.
Official source
宿舍求生：从刷广告开始变强 · 没血条？你管这绝美NPC叫病弱 · 全民躲猫猫：疯批美人她满级天眼
Trace between rank changes and full research
Theme observation · sustained_visibility
Eighties anomaly fiction holds No.11 with unchanged heat
Hongguo AI-drama remains No.11 at 56.03m (5603万). A changed update label is not rising heat. The synopsis includes repairing a television and choosing everyday items; period knowledge should not become an unexplained answer gate.
Hypothesis: inspect an original room, identify a rule conflict and see a causal correction. Without two independent comment sources and competitor searches, this is not a confirmed opportunity.
Official source
八零怀旧怪谈这就是找茬儿
Trace between rank changes and full research
Theme observation · new_in_observed_prefix
Work chat appears as an everyday-boundaries topic
Weibo No.8, heat 109,866. The title expresses tension between personal messaging and work tasks, not demand for an office simulator or evidence of a software change.
Hypothesis: sort urgent, important and rejectable tasks in a fictional inbox with time and relationship feedback. Do not connect real messages; first test whether prioritization is engaging.
Official source
到底谁把微信变成工作软件的
Trace between rank changes and full research
Theme observation · cooling
Crab topic remains at the tail; do not chase a breakout
Douyin remains No.50, with heat 7.667m→7.652m after already reaching the tail last evening. This is a small further decline, not a fresh headline, proof of failure or evidence of crowded supply.
No additional research now. Reconsider original step-based feedback only if independent creators and normal comments establish interest in specific actions; no rebound is promised.
Official source
南北吃蟹大不同
Trace between rank changes and full research
Effective-source alternatives
Source alternative · baidu
The owner most afraid of customers queuing
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · baidu
Young Beijing renters and waste-disposal embarrassment
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · weibo
A finder leaves a question for the owner of RMB 38,000
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · weibo
Who turned WeChat into a work app?
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · douyin
Qipao-inspired cocoa makeup
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · douyin
Autumn outfits in doll collecting
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · bilibili
A superpower to see confession probabilities
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · bilibili
Teaching a younger sister to drive using my voice
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · hongguo
After Wang Ma becomes the Gu family nanny, everyone becomes funny
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · hongguo
Brocade Eighties: Yi He Sends Word North
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · fanqie
Treasure-Gathering Immortal Basin
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research
Source alternative · fanqie
Reborn in an interstellar world, choosing to become a pharmacist
Retained for its concrete setting or action. Normal-comment demand, competitors and remaining reach are unverified; selection does not confirm an opportunity.
Official source
Trace between rank changes and full research