[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-92459":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":13,"contributorsCount":13,"subscribersCount":13,"size":13,"stars1d":13,"stars7d":13,"stars30d":13,"stars90d":13,"forks30d":13,"starsTrendScore":13,"compositeScore":15,"rankGlobal":10,"rankLanguage":10,"license":10,"archived":16,"fork":16,"defaultBranch":17,"hasWiki":18,"hasPages":16,"topics":19,"createdAt":10,"pushedAt":10,"updatedAt":20,"readmeContent":21,"aiSummary":22,"trendingCount":13,"starSnapshotCount":13,"syncStatus":23,"lastSyncTime":24,"discoverSource":25},92459,"neuromind","neuromindapp\u002Fneuromind","neuromindapp","Experimental research system for studying AI forecasting, uncertainty, and evidence-based reasoning. ","https:\u002F\u002Fneuro-sync.app",null,"Python",66,0,1,37,false,"main",true,[],"2026-07-22 04:02:06","# NeuroMind\n\n![NeuroMind banner](frontend\u002Fpublic\u002Fbanner.jpg)\n\n**CA:** `TEZRfLHPRzEQTdSVdxRuAgbuzpKXj9V15KzYYispump`\n\nNeuroMind is a research layer for prediction markets. It scans live Polymarket markets, compares market odds against an independent probability estimate, and turns the disagreement into readable research briefs with source review, rule-risk notes, and a public track record.\n\nThe product is intentionally not a trading bot. It does not custody funds, place bets, or automate market execution. It helps users decide which markets deserve closer study before they act elsewhere.\n\n## Why It Exists\n\nPrediction markets are information-dense but easy to misread. A market can look cheap because the crowd is wrong, because the wording is dangerous, because liquidity is thin, or because the outcome source is ambiguous. NeuroMind separates those questions instead of collapsing them into a single score.\n\nThe useful workflow is:\n\n1. Find markets where the model and the market disagree.\n2. Read the evidence and assumptions behind the estimate.\n3. Check resolution risk before trusting the headline edge.\n4. Compare resolved calls over time through the track record.\n\n## Product Surfaces\n\n### Landing\n\nA minimal dark entry point with the NeuroMind smoke background, clear concept copy, and direct routes into the app or documentation.\n\n### Edge Board\n\nThe research queue. Markets are displayed as compact research cards with:\n\n- live market odds\n- research estimate\n- edge size\n- confidence\n- resolution risk\n- volume\n- locked report preview\n\n### Reports\n\nEach report expands one market into a structured brief:\n\n- market odds versus research estimate\n- key drivers\n- source review\n- rule and settlement risk\n- model notes\n- Polymarket reference link\n\n### Track Record\n\nResolved calls remain visible so the research process can be judged by outcomes, not presentation. The page includes hit rate, confidence bands, simulated points, and individual resolved calls.\n\n### Account\n\nWallet-linked access for report credits, free reports, holder-gated daily access, payment intents, and usage history.\n\n### Docs\n\nStatic documentation for methodology, product surfaces, access model, payments, disclaimers, and privacy notes. Docs deliberately avoid video backgrounds and heavy animation.\n\n## Architecture\n\n```text\nfrontend\u002F     React app served at neuro-mind.app\ndocs\u002F         React docs app served at docs.neuro-mind.app\nbackend\u002F      FastAPI API for auth, reports, markets, quota, payments\nscanner\u002F      Background market scanner worker\ncaddy\u002F        Caddy reverse proxy and static asset server\n```\n\nCore services:\n\n- React + Vite for the app and docs.\n- FastAPI for API routes and report workflows.\n- Postgres for users, markets, reports, payments, and quota usage.\n- Redis for scanner\u002Fruntime coordination.\n- Caddy for TLS, static hosting, and `\u002Fapi` reverse proxying.\n- Privy for wallet login.\n- Helius for Solana holder checks and payment support.\n- Gemini\u002FAnthropic-compatible report generation hooks.\n\n## Local Development\n\nCreate an env file from the example:\n\n```bash\ncp .env.example server.env\n```\n\nInstall and run the frontend:\n\n```bash\ncd frontend\nnpm install\nnpm run dev\n```\n\nInstall and run docs:\n\n```bash\ncd docs\nnpm install\nnpm run dev\n```\n\nRun the full stack with Docker:\n\n```bash\ndocker compose --env-file server.env up -d --build\n```\n\n## Environment\n\nImportant variables:\n\n```text\nDOMAIN=neuro-mind.app\nAPI_ORIGINS=https:\u002F\u002Fneuro-mind.app,https:\u002F\u002Fwww.neuro-mind.app,https:\u002F\u002Fdocs.neuro-mind.app,http:\u002F\u002Flocalhost:5173\nDATABASE_URL=postgresql+asyncpg:\u002F\u002Fneuromind:neuromind@postgres:5432\u002Fneuromind\nREDIS_URL=redis:\u002F\u002Fredis:6379\u002F0\nPRIVY_APP_ID=\nPRIVY_APP_SECRET=\nHELIUS_RPC_URL=\nNEUROMIND_TOKEN_MINT=\nTREASURY_WALLET=\nGEMINI_API_KEY=\nVITE_API_BASE_URL=\u002Fapi\nVITE_PRIVY_APP_ID=\n```\n\nProduction secrets should stay in `server.env` on the server and must not be committed.\n\n## Deployment\n\nThe production stack is built with Docker Compose:\n\n```bash\ndocker compose --env-file server.env up -d --build\n```\n\nCaddy serves:\n\n- `https:\u002F\u002Fneuro-mind.app`\n- `https:\u002F\u002Fwww.neuro-mind.app`\n- `https:\u002F\u002Fdocs.neuro-mind.app`\n\nThe frontend build receives:\n\n- `VITE_API_BASE_URL=\u002Fapi`\n- `VITE_PRIVY_APP_ID=${PRIVY_APP_ID}`\n\n## Research Model\n\nNeuroMind treats a market estimate as a research artifact, not a command. Reports should explain:\n\n- what the market is pricing\n- what the model estimates\n- why the gap exists\n- what evidence could break the thesis\n- whether the market wording can resolve unexpectedly\n\nThe track record is part of the product because probability work should be judged after resolution.\n\n## Safety Boundaries\n\nNeuroMind is research software only.\n\n- It does not provide financial advice.\n- It does not guarantee outcomes.\n- It does not trade on behalf of users.\n- It does not custody market funds.\n- It is not affiliated with Polymarket.\n\nUsers are responsible for reading the market rules, understanding their jurisdiction, and making their own decisions.\n","NeuroMind 是一个面向预测市场的研究型分析系统，用于识别市场共识与独立概率模型之间的分歧，并生成结构化研究简报。其核心功能包括实时扫描 Polymarket 市场、生成基于证据的概率估计、评估分辨率风险与规则风险、提供可追溯的公开实绩记录。技术上采用 React\u002FVite 前端、FastAPI 后端、PostgreSQL 存储及 Gemini\u002FAnthropic 兼容模型生成报告，强调人工决策支持而非自动交易。适用于预测市场研究者、信息套利分析人员及需要审慎评估事件概率与结算不确定性的专业用户。",2,"2026-07-09 02:30:05","CREATED_QUERY"]