[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-98312":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":11,"openIssues":12,"contributorsCount":12,"subscribersCount":12,"size":12,"stars1d":12,"stars7d":12,"stars30d":12,"stars90d":12,"forks30d":12,"starsTrendScore":12,"compositeScore":13,"rankGlobal":9,"rankLanguage":9,"license":14,"archived":15,"fork":15,"defaultBranch":16,"hasWiki":17,"hasPages":15,"topics":18,"createdAt":9,"pushedAt":19,"updatedAt":20,"readmeContent":9,"aiSummary":9,"trendingCount":12,"starSnapshotCount":12,"syncStatus":21,"lastSyncTime":22,"discoverSource":23},98312,"Grader-agent","erictrainnlpinxihu\u002FGrader-agent","erictrainnlpinxihu","Grader 的 agent 采用五阶段主循环：perceive→plan→act→observe→respond，由 harness 驱动。模型只决定意图与置信度，路由、工具集、风险等级由代码钉死。act 分只读 ReAct、RAG、人工提案、批量初批。四个写动作不在工具表，只能产提案。批改输出结构化 GradingDraft，逐条绑定 rubric 与段落 hash，保证可重放。",null,"Python",101,0,36,"MIT License",false,"main",true,[],"2026-09-28 15:39:27","2026-10-09 12:43:45",2,"2026-10-13 04:09:46","CREATED_QUERY"]