[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-automate-web3-grant-screening-ai-scoring-zh":3,"article-related-automate-web3-grant-screening-ai-scoring-zh":30,"series-ai-agent-ee3c0a0e-0117-4f79-b94b-308d08b43669":77},{"id":4,"slug":5,"title":6,"content":7,"summary":8,"source":9,"source_url":10,"author":11,"image_url":12,"cover_image":12,"category":13,"language":14,"translated_content":11,"related_article_id":15,"keywords":16,"key_takeaways":22,"views":26,"created_at":27,"published_at":28,"topic_cluster_id":29},"ee3c0a0e-0117-4f79-b94b-308d08b43669","automate-web3-grant-screening-ai-scoring-zh","Web3 補助金 AI 篩選流程實作","\u003Cp>想把 \u003Ca href=\"\u002Ftag\u002Fweb3\">Web3\u003C\u002Fa> 補助金審查自動化，但又不想失去人工把關嗎？\u003C\u002Fp>\u003Cp data-speakable=\"summary\">這篇教你建立一個有人審核的 Web3 補助金 AI 篩選流程。\u003C\u002Fp>\u003Cp>這份操作指南適合要處理提案量、又必須保留公平性、可稽核性與人工責任的團隊。照著做完，你會得到一套可執行的審查流程，能先檢查資格、再依評分表打分、標記可疑案例，最後把不確定案件交給審查者。\u003C\u002Fp>\u003Ch2>開始之前\u003C\u002Fh2>\u003Cul>\u003Cli>一個可用的 GitHub 帳號與補助金專案儲存庫。\u003C\u002Fli>\u003Cli>LLM 供應商、區塊鏈資料來源、身分工具的 API key。\u003C\u002Fli>\u003Cli>Node 20+ 或 Python 3.11+，用來執行篩選服務。\u003C\u002Fli>\u003Cli>一份帶權重的補助金評分表，例如影響力、可行性、生態契合度。\u003C\u002Fli>\u003Cli>申請文件、錢包地址、鏈上活動資料的存取權。\u003C\u002Fli>\u003Cli>具名審查者與稽核紀錄目的地的人工覆核流程。\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>Step 1: 定義資格規則\u003C\u002Fh2>\u003Cp>目的：先建立硬性門檻，讓 AI 在打分前就能一致地檢查每份申請。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1784187173996-8c2e.png\" alt=\"Web3 補助金 AI 篩選流程實作\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>把錢包年齡、專案類別、必填文件、預算範圍、過往補助紀錄與生態契合度寫成明確\u003Ca href=\"\u002Fnews\u002Fanthropic-state-ai-rules-crypto-impact-zh\">規則\u003C\u002Fa>，並確保它們可以被機器直接判讀。規則越具體，後面的自動化越不會變成猜測。\u003C\u002Fp>\u003Cp>驗收：你應該能在不讀完整提案的情況下，直接拒絕缺文件或超出範圍的申請。\u003C\u002Fp>\u003Ch2>Step 2: 加入身分與 Sybil 檢查\u003C\u002Fh2>\u003Cp>目的：在任何評分之前，先確認每份申請對應到真實且唯一的參與者。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1784187182512-3xyi.png\" alt=\"Web3 補助金 AI 篩選流程實作\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>結合 proof-of-personhood、passport 式身分訊號、錢包歷史與重複偵測，先抓出重複送件與協同行為。這一層要先跑，因為假身分拿到高分，結果仍然是壞決策。\u003C\u002Fp>\u003Cp>驗收：你應該在產生評分前，就看到重複錢包、重複中繼\u003Ca href=\"\u002Fnews\u002Fmetaperch-metadata-bioacoustics-foundation-models-zh\">資料\u003C\u002Fa>或低信任身分被標記。\u003C\u002Fp>\u003Ch2>Step 3: 依公開評分表產生分數\u003C\u002Fh2>\u003Cp>目的：把主觀審查轉成可重複的評分流程，並為每個分數保留證據。\u003C\u002Fp>\u003Cp>讓模型只根據你公開給申請者的評分表打分，並要求引用提案文字或連結文件中的原句。實用的評分表通常包含技術性、影響力、透明度、社群價值與可行性。\u003C\u002Fp>\u003Cpre>\u003Ccode>rubric = {\n  \"technical_merit\": 0.30,\n  \"ecosystem_alignment\": 0.25,\n  \"feasibility\": 0.20,\n  \"transparency\": 0.15,\n  \"community_value\": 0.10\n}\n\nprompt = \"Score each category from 1-5 and cite the exact evidence used.\"\n\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>驗收：你應該看到一份結構化的分數卡，包含各分類分數與引用證據，而不是自由格式的 yes\u002Fno 回答。\u003C\u002Fp>\u003Ch2>Step 4: 轉成風險標記\u003C\u002Fh2>\u003Cp>目的：讓 AI 保持顧問角色，輸出風險訊號，而不是直接做最終核准。\u003C\u002Fp>\u003Cp>把系統設定成輸出風險分數與標記，例如可能重複、文件不完整、鏈上活動異常或里程碑敘述不一致。中等信心案件應送人工，高風險與低風險案件則可用來加速分流。\u003C\u002Fp>\u003Cp>驗收：你應該看到一個審查佇列，而模型結果只是決策輸入之一，不是決策本身。\u003C\u002Fp>\u003Ch2>Step 5: 記錄審查並回寫結果\u003C\u002Fh2>\u003Cp>目的：建立稽核軌跡與回饋迴圈，讓每一輪補助金審查都能持續改善。\u003C\u002Fp>\u003Cp>保存使用的來源\u003Ca href=\"\u002Fnews\u002Fmojo-unlabeled-training-neural-decoding-zh\">資料\u003C\u002Fa>、觸發規則或標記、模型版本、核准或覆寫的審查者，以及最終結果。接著比較審查者分歧、里程碑完成率與預算準確度，調整提示詞或重新訓練模型。\u003C\u002Fp>\u003Cp>驗收：你應該能重建任何一筆申請為什麼被標記，也能看出下一輪流程是否更好。\u003C\u002Fp>\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>指標\u003C\u002Fth>\u003Cth>基準／優化前\u003C\u002Fth>\u003Cth>結果／優化後\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>審查範圍\u003C\u002Ftd>\u003Ctd>逐份人工閱讀所有申請\u003C\u002Ftd>\u003Ctd>AI 先篩掉明顯重複與缺件案件\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>決策型態\u003C\u002Ftd>\u003Ctd>臨時性的通過／拒絕判斷\u003C\u002Ftd>\u003Ctd>結構化風險分數加上審查標記\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>可稽核性\u003C\u002Ftd>\u003Ctd>只有零散備註與 email 往來\u003C\u002Ftd>\u003Ctd>記錄模型版本、證據與審查者動作\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>人工監督\u003C\u002Ftd>\u003Ctd>升級處理不一致\u003C\u002Ftd>\u003Ctd>中等信心案件自動分流給審查者\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003Ch2>常見錯誤\u003C\u002Fh2>\u003Cul>\u003Cli>把 AI 當成最後決策者。修法：保留人工核准權，讓模型只做建議與分流。\u003C\u002Fli>\u003Cli>讓申請者影響提示詞。修法：清理隱藏文字、過濾附件內容，並把申請資料視為不可信輸入。\u003C\u002Fli>\u003Cli>跳過稽核紀錄。修法：為每個案件記錄輸入、輸出、審查者動作與模型版本。\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>接下來可以看什麼\u003C\u002Fh2>\u003Cp>當你的篩選流程穩定後，可以再加入申訴處理、偏差檢查與定期政策回顧，讓補助金流程在申請量成長時仍然公平可控。\u003C\u002Fp>","建立一個有人審核的 AI 補助金篩選流程，先做資格檢查，再依公開評分表打分、標記風險，最後把結果送給人工覆核。","financefeeds.com","https:\u002F\u002Ffinancefeeds.com\u002Fhow-to-set-up-an-ai-assisted-evaluation-protocol-for-autonomous-web3-grant-screening\u002F",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1784187173996-8c2e.png","ai-agent","zh","ae8b2df7-05be-4377-8fa8-00856625839e",[17,18,19,20,21],"Web3","補助金審查","AI scoring","LLM","Sybil 檢查",[23,24,25],"先用硬性資格規則與身分檢查擋掉不合格案件。","評分要依公開 rubric，並要求模型引用證據。","AI 只做風險分流與建議，最終決策保留給人工。",0,"2026-07-16T07:32:23.745459+00:00","2026-07-16T07:32:23.717+00:00","e3b68196-9e64-4c18-a3b6-a73e73bfb367",{"tags":31,"relatedLang":36,"relatedPosts":40},[32,34],{"name":20,"slug":33},"llm",{"name":17,"slug":35},"web3",{"id":15,"slug":37,"title":38,"language":39},"automate-web3-grant-screening-ai-scoring-en","Automate Web3 Grant Screening With AI Scoring","en",[41,47,53,59,65,71],{"id":42,"slug":43,"title":44,"cover_image":45,"image_url":45,"created_at":46,"category":13},"02bfe363-1ca6-4c20-be08-34d4ad532b71","anthropic-model-task-persistence-tuning-zh","Anthropic 任務耐力調校指南","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1784102587342-bz4a.png","2026-07-15T08:02:31.959666+00:00",{"id":48,"slug":49,"title":50,"cover_image":51,"image_url":51,"created_at":52,"category":13},"7acb5b4c-de5e-4ded-83cf-82cf93f47a00","google-gemini-enterprise-agent-platform-cloud-service-zh","Google Gemini Enterprise 代理平台把 AI 代理變成雲…","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1784059372531-sfab.png","2026-07-14T20:02:24.962585+00:00",{"id":54,"slug":55,"title":56,"cover_image":57,"image_url":57,"created_at":58,"category":13},"1f24f862-f37b-4bd9-b78d-434713905348","workbuddy-harness-engineering-agent-reliability-zh","WorkBuddy 證明了 Agent 可靠性不靠大模型本身","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1783987377305-yt32.png","2026-07-14T00:02:31.385926+00:00",{"id":60,"slug":61,"title":62,"cover_image":63,"image_url":63,"created_at":64,"category":13},"415418c7-749e-4352-a21d-d2fa62d8b96b","perplexity-teammate-coding-agent-strategy-zh","Perplexity 應把 Teammate 做成 coding agent，…","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1783589576714-n7gs.png","2026-07-09T09:32:25.568287+00:00",{"id":66,"slug":67,"title":68,"cover_image":69,"image_url":69,"created_at":70,"category":13},"884f1bb8-4bae-4cfa-87d1-b323be1d6166","hp-adopts-openai-frontier-global-operations-zh","HP 將 Frontier 送進全球營運","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1783362776958-1e1n.png","2026-07-06T18:32:30.897068+00:00",{"id":72,"slug":73,"title":74,"cover_image":75,"image_url":75,"created_at":76,"category":13},"b97b8932-56a9-431f-8270-3f892f8feb94","build-production-vector-db-rag-pipeline-zh","用 n8n 建出可上線的向量資料庫","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1783258377362-cg5x.png","2026-07-05T13:32:23.466634+00:00",[78,83,88,93,98,103,108,113,118,123],{"id":79,"slug":80,"title":81,"created_at":82},"4ae1e197-1d3d-4233-8733-eafe9cb6438b","claude-now-uses-your-pc-to-finish-tasks-zh","Claude 開始幫你操作電腦","2026-03-26T07:20:48.457387+00:00",{"id":84,"slug":85,"title":86,"created_at":87},"5bede67f-e21c-413d-9ab8-54a3c3d26227","googles-2026-ai-agent-report-decoded-zh","Google 2026 AI Agent 報告解讀","2026-03-26T11:15:22.651956+00:00",{"id":89,"slug":90,"title":91,"created_at":92},"2987d097-563f-46c7-b76f-b558d8ef7c2b","kimi-k25-review-stronger-still-not-legend-zh","Kimi K2.5 評測：更強，但還不是神作","2026-03-27T07:15:55.277513+00:00",{"id":94,"slug":95,"title":96,"created_at":97},"95c9053b-e3f4-4cb5-aace-5c54f4c9e044","claude-code-controls-mac-desktop-zh","Claude Code 也能操控 Mac 了","2026-03-28T03:01:58.58121+00:00",{"id":99,"slug":100,"title":101,"created_at":102},"dc58e153-e3a8-4c06-9b96-1aa64eabbf5f","cloudflare-100x-faster-ai-agent-sandbox-zh","Cloudflare 的 AI 沙箱跑超快","2026-03-28T03:09:44.142236+00:00",{"id":104,"slug":105,"title":106,"created_at":107},"1c8afc56-253f-47a2-979f-1065ff072f2a","openai-backs-isara-agent-swarm-bet-zh","OpenAI 挺 Isara 的 agent swarm …","2026-03-28T03:15:27.513155+00:00",{"id":109,"slug":110,"title":111,"created_at":112},"7379b422-576e-45df-ad5a-d57a0d9dd467","openai-plan-automated-ai-researcher-zh","OpenAI 想做自動化 AI 研究員","2026-03-28T03:17:42.090548+00:00",{"id":114,"slug":115,"title":116,"created_at":117},"48c9889e-86df-450b-a356-e4a4b7c83c5b","harness-engineering-ai-agent-reliability-2026-zh","駕馭工程：從「馬具」到「作業系統」，AI Agent 可靠性的終極密碼","2026-03-31T06:42:53.556721+00:00",{"id":119,"slug":120,"title":121,"created_at":122},"96d8e8c8-1edd-475d-9145-b1e7a1b02b65","mcp-explained-from-prompts-to-production-zh","MCP 怎麼把提示詞變工作流","2026-04-01T09:24:39.321274+00:00",{"id":124,"slug":125,"title":126,"created_at":127},"f2ca7720-b471-4ce5-9336-2a9ac2a876fd","amazon-bedrock-agents-multi-agent-workflows-zh","Amazon Bedrock Agents 進入多代理工作流","2026-04-01T09:30:29.945429+00:00"]