{"generatedAt":"2026-04-16T16:18:33.661849+00:00","errors":[],"all":[{"title":"NousResearch/hermes-agent — The agent that grows with you","url":"https://github.com/NousResearch/hermes-agent","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":6425,"repoAgeDays":267.7,"stars":92960,"forks":12878,"watchers":375,"openIssues":4942,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":816,"recentCommits30d":2164,"tags":["other"],"score":21.072,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":18.751,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":18.751,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":18.751,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.639,"recent_star_ratio":0.344,"recent_commits":9.491,"network":4.522,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":3,"rank_delta":2,"image":"https://opengraph.githubassets.com/1/NousResearch/hermes-agent","rank":1},{"title":"langgenius/dify — Production-ready platform for agentic workflow development.","url":"https://github.com/langgenius/dify","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":26408,"repoAgeDays":1100.3,"stars":138010,"forks":21632,"watchers":780,"openIssues":940,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":272,"recentCommits30d":828,"tags":["AI"],"score":21.037,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":17.565,"novelty":1.025,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.13,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":17.565,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":17.565,"novelty":1.025,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.574,"recent_star_ratio":0.232,"recent_commits":8.073,"network":4.815,"confidence":1.0,"truncation_penalty":0.58}}},"rank_prev":1,"rank_delta":-1,"image":"https://opengraph.githubassets.com/1/langgenius/dify","rank":2},{"title":"anomalyco/opencode — The open source coding agent.","url":"https://github.com/anomalyco/opencode","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":8420,"repoAgeDays":350.8,"stars":144334,"forks":16313,"watchers":570,"openIssues":6090,"defaultBranch":"dev","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":365,"recentCommits30d":999,"tags":["other"],"score":20.523,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":17.795,"novelty":0.366,"confirmation":0.042,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":1.1,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":17.795,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":17.795,"novelty":0.366,"confirmation":0.042,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.568,"recent_star_ratio":0.222,"recent_commits":8.421,"network":4.702,"confidence":1.0,"truncation_penalty":0.58}}},"rank_prev":9,"rank_delta":6,"image":"https://opengraph.githubassets.com/1/anomalyco/opencode","rank":3},{"title":"openai/codex — Lightweight coding agent that runs in your terminal","url":"https://github.com/openai/codex","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":8842,"repoAgeDays":368.4,"stars":75671,"forks":10738,"watchers":430,"openIssues":2708,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":215,"recentCommits30d":844,"tags":["AI"],"score":20.433,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":17.135,"novelty":0.935,"confirmation":0.042,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":1.77,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":17.135,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":17.135,"novelty":0.935,"confirmation":0.042,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.685,"recent_star_ratio":0.423,"recent_commits":7.87,"network":4.482,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":7,"rank_delta":3,"image":"https://opengraph.githubassets.com/1/openai/codex","rank":4},{"title":"google-gemini/gemini-cli — An open-source AI agent that brings the power of Gemini directly into your terminal.","url":"https://github.com/google-gemini/gemini-cli","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":8735,"repoAgeDays":364.0,"stars":101459,"forks":13151,"watchers":538,"openIssues":2958,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":86,"recentCommits30d":539,"tags":["AI"],"score":19.558,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":16.176,"novelty":0.935,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":2.01,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":16.176,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":16.176,"novelty":0.935,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.623,"recent_star_ratio":0.315,"recent_commits":6.851,"network":4.602,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":10,"rank_delta":5,"image":"https://opengraph.githubassets.com/1/google-gemini/gemini-cli","rank":5},{"title":"code-yeongyu/oh-my-openagent — omo; the best agent harness - previously oh-my-opencode","url":"https://github.com/code-yeongyu/oh-my-openagent","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":3230,"repoAgeDays":134.6,"stars":52082,"forks":4186,"watchers":199,"openIssues":533,"defaultBranch":"dev","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":246,"recentCommits30d":1263,"tags":["other"],"score":19.472,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":17.151,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":17.151,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":17.151,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.796,"recent_star_ratio":0.614,"recent_commits":8.172,"network":4.092,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":5,"rank_delta":-1,"image":"https://opengraph.githubassets.com/1/code-yeongyu/oh-my-openagent","rank":6},{"title":"ray-project/ray — Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.","url":"https://github.com/ray-project/ray","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":83036,"repoAgeDays":3459.8,"stars":42157,"forks":7447,"watchers":479,"openIssues":3579,"defaultBranch":"master","citations":0,"starsPerDay":17.7,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":133,"recentStars30d":531,"recentStarsPerDay7d":19.0,"stargazerHistoryTruncated":false,"recentCommits7d":102,"recentCommits30d":411,"tags":["other"],"score":19.413,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":17.091,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":17.091,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":17.091,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":4.774,"recent_star_ratio":0.202,"recent_commits":6.881,"network":4.351,"confidence":0.999,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/ray-project/ray","rank":7},{"title":"HKUDS/nanobot — \"🐈 nanobot: The Ultra-Lightweight Personal AI Agent\"","url":"https://github.com/HKUDS/nanobot","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":1784,"repoAgeDays":74.3,"stars":39783,"forks":6982,"watchers":197,"openIssues":940,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":143,"recentCommits30d":609,"tags":["AI"],"score":18.842,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":16.521,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":16.521,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":16.521,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.906,"recent_star_ratio":0.804,"recent_commits":7.359,"network":4.17,"confidence":0.998,"truncation_penalty":0.58}}},"rank_prev":8,"rank_delta":0,"image":"https://opengraph.githubassets.com/1/HKUDS/nanobot","rank":8},{"title":"affaan-m/everything-claude-code — The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.","url":"https://github.com/affaan-m/everything-claude-code","source":"github.com","categories":["repo"],"publishedHoursAgo":13,"updatedHoursAgo":13,"createdHoursAgo":2127,"repoAgeDays":88.6,"stars":158332,"forks":24617,"watchers":828,"openIssues":110,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":183,"recentCommits30d":759,"tags":["other"],"score":18.721,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.973,"momentum":16.833,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.15,"buzz":0.0,"hype":0.25,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.973,"velocity":16.833,"score_breakdown":{"trust":0.765,"freshness":0.973,"momentum":16.833,"novelty":0.0,"confirmation":0.0,"spike":0.15,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.733,"recent_stars":3.557,"recent_star_ratio":0.202,"recent_commits":7.678,"network":4.871,"confidence":1.0,"truncation_penalty":0.58}}},"rank_prev":12,"rank_delta":3,"image":"https://opengraph.githubassets.com/1/affaan-m/everything-claude-code","rank":9},{"title":"bytedance/deer-flow — An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.","url":"https://github.com/bytedance/deer-flow","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":8269,"repoAgeDays":344.5,"stars":62006,"forks":7995,"watchers":272,"openIssues":654,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":50,"recentCommits30d":297,"tags":["AI"],"score":18.627,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":15.248,"novelty":0.931,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.02,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":15.248,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":15.248,"novelty":0.931,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.739,"recent_star_ratio":0.516,"recent_commits":6.102,"network":4.311,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":15,"rank_delta":5,"image":"https://opengraph.githubassets.com/1/bytedance/deer-flow","rank":10},{"title":"shanraisshan/claude-code-best-practice — from vibe coding to agentic engineering - practice makes claude perfect","url":"https://github.com/shanraisshan/claude-code-best-practice","source":"github.com","categories":["repo"],"publishedHoursAgo":9,"updatedHoursAgo":9,"createdHoursAgo":4009,"repoAgeDays":167.0,"stars":45449,"forks":4388,"watchers":357,"openIssues":10,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":88,"recentCommits30d":217,"tags":["AI"],"score":18.53,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.051,"momentum":15.323,"novelty":1.025,"confirmation":0.126,"ran_strength":0.0,"spike":0.24,"buzz":0.12,"hype":1.98,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.051,"velocity":15.323,"score_breakdown":{"trust":0.765,"freshness":1.051,"momentum":15.323,"novelty":1.025,"confirmation":0.126,"spike":0.24,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.842,"recent_stars":3.848,"recent_star_ratio":0.704,"recent_commits":6.463,"network":4.191,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":14,"rank_delta":3,"image":"https://opengraph.githubassets.com/1/shanraisshan/claude-code-best-practice","rank":11},{"title":"MemPalace/mempalace — The best-benchmarked open-source AI memory system. And it's free.","url":"https://github.com/MemPalace/mempalace","source":"github.com","categories":["repo"],"publishedHoursAgo":14,"updatedHoursAgo":14,"createdHoursAgo":278,"repoAgeDays":11.6,"stars":47008,"forks":6138,"watchers":289,"openIssues":381,"defaultBranch":"develop","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":249,"recentCommits30d":407,"tags":["AI"],"score":18.296,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.955,"momentum":16.426,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.15,"buzz":0.0,"hype":0.25,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.955,"velocity":16.426,"score_breakdown":{"trust":0.765,"freshness":0.955,"momentum":16.426,"novelty":0.0,"confirmation":0.0,"spike":0.15,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.708,"recent_stars":3.834,"recent_star_ratio":0.681,"recent_commits":7.674,"network":4.23,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":6,"rank_delta":-6,"image":"https://opengraph.githubassets.com/1/MemPalace/mempalace","rank":12},{"title":"aaif-goose/goose — an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM","url":"https://github.com/aaif-goose/goose","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":14421,"repoAgeDays":600.9,"stars":42302,"forks":4266,"watchers":216,"openIssues":333,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":60,"recentCommits30d":277,"tags":["AI"],"score":18.014,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":15.285,"novelty":0.366,"confirmation":0.042,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":1.1,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":15.285,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":15.285,"novelty":0.366,"confirmation":0.042,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.878,"recent_star_ratio":0.756,"recent_commits":6.232,"network":4.086,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/aaif-goose/goose","rank":13},{"title":"hesreallyhim/awesome-claude-code — A curated list of awesome skills, hooks, slash-commands, agent orchestrators, applications, and plugins for Claude Code by Anthropic","url":"https://github.com/hesreallyhim/awesome-claude-code","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":8683,"repoAgeDays":361.8,"stars":39089,"forks":3213,"watchers":260,"openIssues":220,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":61,"recentCommits30d":189,"tags":["other"],"score":17.951,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":15.125,"novelty":0.485,"confirmation":0.021,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":1.18,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":15.125,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":15.125,"novelty":0.485,"confirmation":0.021,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.914,"recent_star_ratio":0.819,"recent_commits":6.076,"network":4.047,"confidence":0.998,"truncation_penalty":0.58}}},"rank_prev":11,"rank_delta":-3,"image":"https://opengraph.githubassets.com/1/hesreallyhim/awesome-claude-code","rank":14},{"title":"badlogic/pi-mono — AI agent toolkit: coding agent CLI, unified LLM API, TUI & web UI libraries, Slack bot, vLLM pods","url":"https://github.com/badlogic/pi-mono","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":6002,"repoAgeDays":250.1,"stars":36287,"forks":4200,"watchers":146,"openIssues":49,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":81,"recentCommits30d":322,"tags":["AI"],"score":17.917,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":15.596,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":15.596,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":15.596,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.951,"recent_star_ratio":0.882,"recent_commits":6.566,"network":3.994,"confidence":0.998,"truncation_penalty":0.58}}},"rank_prev":13,"rank_delta":-2,"image":"https://opengraph.githubassets.com/1/badlogic/pi-mono","rank":15},{"title":"OpenHands/OpenHands — 🙌 OpenHands: AI-Driven Development","url":"https://github.com/OpenHands/OpenHands","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":18348,"repoAgeDays":764.5,"stars":71319,"forks":8968,"watchers":434,"openIssues":413,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":464,"recentStars30d":800,"recentStarsPerDay7d":66.29,"stargazerHistoryTruncated":true,"recentCommits7d":53,"recentCommits30d":231,"tags":["other"],"score":17.282,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":14.961,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":14.961,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":14.961,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.385,"recent_star_ratio":0.298,"recent_commits":6.041,"network":4.437,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/OpenHands/OpenHands","rank":16},{"title":"crewAIInc/crewAI — Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.","url":"https://github.com/crewAIInc/crewAI","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":21660,"repoAgeDays":902.5,"stars":49021,"forks":6700,"watchers":352,"openIssues":525,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":631,"recentStars30d":800,"recentStarsPerDay7d":90.14,"stargazerHistoryTruncated":true,"recentCommits7d":42,"recentCommits30d":210,"tags":["AI"],"score":17.152,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":14.831,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":14.831,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":14.831,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.655,"recent_star_ratio":0.542,"recent_commits":5.793,"network":4.29,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/crewAIInc/crewAI","rank":17},{"title":"mem0ai/mem0 — Universal memory layer for AI Agents","url":"https://github.com/mem0ai/mem0","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":24751,"repoAgeDays":1031.3,"stars":53229,"forks":5966,"watchers":225,"openIssues":229,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":37,"recentCommits30d":194,"tags":["AI"],"score":17.045,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":14.723,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":14.723,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":14.723,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.788,"recent_star_ratio":0.601,"recent_commits":5.647,"network":4.194,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/mem0ai/mem0","rank":18},{"title":"thedotmack/claude-mem — A Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude's agent-sdk), and injects relevant context back into future sessions.","url":"https://github.com/thedotmack/claude-mem","source":"github.com","categories":["repo"],"publishedHoursAgo":17,"updatedHoursAgo":17,"createdHoursAgo":5467,"repoAgeDays":227.8,"stars":59073,"forks":4828,"watchers":226,"openIssues":140,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":58,"recentCommits30d":247,"tags":["AI"],"score":16.505,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.901,"momentum":14.689,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.15,"buzz":0.0,"hype":0.25,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.901,"velocity":14.689,"score_breakdown":{"trust":0.765,"freshness":0.901,"momentum":14.689,"novelty":0.0,"confirmation":0.0,"spike":0.15,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.638,"recent_stars":3.754,"recent_star_ratio":0.542,"recent_commits":6.151,"network":4.161,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":21,"rank_delta":2,"image":"https://opengraph.githubassets.com/1/thedotmack/claude-mem","rank":19},{"title":"FlowiseAI/Flowise — Build AI Agents, Visually","url":"https://github.com/FlowiseAI/Flowise","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":26691,"repoAgeDays":1112.1,"stars":51976,"forks":24163,"watchers":355,"openIssues":814,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":304,"recentStars30d":800,"recentStarsPerDay7d":43.43,"stargazerHistoryTruncated":true,"recentCommits7d":25,"recentCommits30d":113,"tags":["AI"],"score":16.276,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":13.954,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":13.954,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":13.954,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.215,"recent_star_ratio":0.31,"recent_commits":5.064,"network":4.581,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/FlowiseAI/Flowise","rank":20},{"title":"ChromeDevTools/chrome-devtools-mcp — Chrome DevTools for coding agents","url":"https://github.com/ChromeDevTools/chrome-devtools-mcp","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":5213,"repoAgeDays":217.2,"stars":35438,"forks":2157,"watchers":133,"openIssues":94,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":25,"recentCommits30d":87,"tags":["other"],"score":16.147,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":13.825,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":13.825,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":13.825,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.963,"recent_star_ratio":0.903,"recent_commits":4.947,"network":3.828,"confidence":0.998,"truncation_penalty":0.58}}},"rank_prev":18,"rank_delta":-3,"image":"https://opengraph.githubassets.com/1/ChromeDevTools/chrome-devtools-mcp","rank":21},{"title":"obra/superpowers — An agentic skills framework & software development methodology that works.","url":"https://github.com/obra/superpowers","source":"github.com","categories":["repo"],"publishedHoursAgo":14,"updatedHoursAgo":14,"createdHoursAgo":4532,"repoAgeDays":188.8,"stars":155820,"forks":13524,"watchers":644,"openIssues":296,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":12,"recentCommits30d":53,"tags":["AI"],"score":16.078,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.955,"momentum":13.057,"novelty":1.025,"confirmation":0.126,"ran_strength":0.0,"spike":0.15,"buzz":0.12,"hype":1.83,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.955,"velocity":13.057,"score_breakdown":{"trust":0.765,"freshness":0.955,"momentum":13.057,"novelty":1.025,"confirmation":0.126,"spike":0.15,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.708,"recent_stars":3.559,"recent_star_ratio":0.205,"recent_commits":4.103,"network":4.692,"confidence":1.0,"truncation_penalty":0.58}}},"rank_prev":23,"rank_delta":1,"image":"https://opengraph.githubassets.com/1/obra/superpowers","rank":22},{"title":"CherryHQ/cherry-studio — AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":16622,"repoAgeDays":692.6,"stars":43607,"forks":4094,"watchers":189,"openIssues":1002,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"recentStars7d":459,"recentStars30d":800,"recentStarsPerDay7d":65.57,"stargazerHistoryTruncated":true,"recentCommits7d":21,"recentCommits30d":158,"tags":["AI"],"score":16.02,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.226,"momentum":13.699,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":13.699,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":13.699,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.488,"recent_star_ratio":0.484,"recent_commits":5.063,"network":4.056,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/CherryHQ/cherry-studio","rank":23},{"title":"msitarzewski/agency-agents — A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.","url":"https://github.com/msitarzewski/agency-agents","source":"github.com","categories":["repo"],"publishedHoursAgo":107,"updatedHoursAgo":107,"createdHoursAgo":4444,"repoAgeDays":185.2,"stars":80920,"forks":12956,"watchers":628,"openIssues":91,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":36,"recentCommits30d":65,"tags":["AI"],"score":14.346,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.159,"momentum":13.422,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.0,"buzz":0.0,"hype":0.0,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.159,"velocity":13.422,"score_breakdown":{"trust":0.765,"freshness":0.159,"momentum":13.422,"novelty":0.0,"confirmation":0.0,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.028,"recent_stars":3.669,"recent_star_ratio":0.395,"recent_commits":5.135,"network":4.599,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":19,"rank_delta":-5,"image":"https://opengraph.githubassets.com/1/msitarzewski/agency-agents","rank":24},{"title":"asgeirtj/system_prompts_leaks — Extracted system prompts from ChatGPT (GPT-5.4, GPT-5.3, Codex), Claude (Opus 4.6, Sonnet 4.6, Claude Code), Gemini (3.1 Pro, 3 Flash, CLI), Grok (4.2, 4), Perplexity, and more. Updated regularly.","url":"https://github.com/asgeirtj/system_prompts_leaks","source":"github.com","categories":["repo"],"publishedHoursAgo":62,"updatedHoursAgo":62,"createdHoursAgo":8365,"repoAgeDays":348.5,"stars":38436,"forks":6325,"watchers":496,"openIssues":19,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":466,"recentStars30d":800,"recentStarsPerDay7d":66.57,"stargazerHistoryTruncated":true,"recentCommits7d":14,"recentCommits30d":40,"tags":["other"],"score":14.274,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.379,"momentum":12.069,"novelty":0.935,"confirmation":0.126,"ran_strength":0.0,"spike":0.0,"buzz":0.0,"hype":1.46,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.379,"velocity":12.069,"score_breakdown":{"trust":0.765,"freshness":0.379,"momentum":12.069,"novelty":0.935,"confirmation":0.126,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.134,"recent_stars":3.535,"recent_star_ratio":0.554,"recent_commits":4.108,"network":4.31,"confidence":0.998,"truncation_penalty":0.58}}},"rank_prev":20,"rank_delta":-5,"image":"https://opengraph.githubassets.com/1/asgeirtj/system_prompts_leaks","rank":25},{"title":"VoltAgent/awesome-design-md — A collection of DESIGN.md files inspired by popular brand design systems. Drop one into your project and let coding agents generate a matching UI.","url":"https://github.com/VoltAgent/awesome-design-md","source":"github.com","categories":["repo"],"publishedHoursAgo":7,"updatedHoursAgo":7,"createdHoursAgo":383,"repoAgeDays":16.0,"stars":55886,"forks":6913,"watchers":232,"openIssues":267,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":4,"recentCommits30d":32,"tags":["other"],"score":14.018,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.092,"momentum":11.921,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.24,"buzz":0.0,"hype":0.4,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.092,"velocity":11.921,"score_breakdown":{"trust":0.765,"freshness":1.092,"momentum":11.921,"novelty":0.0,"confirmation":0.0,"spike":0.24,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.902,"recent_stars":3.772,"recent_star_ratio":0.573,"recent_commits":3.022,"network":4.238,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":26,"rank_delta":-1,"image":"https://opengraph.githubassets.com/1/VoltAgent/awesome-design-md","rank":27},{"title":"anthropics/skills — Public repository for Agent Skills","url":"https://github.com/anthropics/skills","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":4944,"repoAgeDays":206.0,"stars":118783,"forks":13722,"watchers":810,"openIssues":710,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":2,"recentCommits30d":5,"tags":["other"],"score":14.027,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":11.2,"novelty":0.485,"confirmation":0.021,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":1.18,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":11.2,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":11.2,"novelty":0.485,"confirmation":0.021,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":3.596,"recent_star_ratio":0.269,"recent_commits":1.795,"network":4.704,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":27,"rank_delta":1,"image":"https://opengraph.githubassets.com/1/anthropics/skills","rank":26},{"title":"google/langextract — A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.","url":"https://github.com/google/langextract","source":"github.com","categories":["repo"],"publishedHoursAgo":7,"updatedHoursAgo":7,"createdHoursAgo":6763,"repoAgeDays":281.8,"stars":35662,"forks":2424,"watchers":160,"openIssues":117,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":137,"recentStars30d":800,"recentStarsPerDay7d":19.57,"stargazerHistoryTruncated":true,"recentCommits7d":4,"recentCommits30d":14,"tags":["other"],"score":12.415,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.092,"momentum":10.318,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.24,"buzz":0.0,"hype":0.4,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.092,"velocity":10.318,"score_breakdown":{"trust":0.765,"freshness":1.092,"momentum":10.318,"novelty":0.0,"confirmation":0.0,"spike":0.24,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.902,"recent_stars":2.879,"recent_star_ratio":0.302,"recent_commits":2.667,"network":3.887,"confidence":0.998,"truncation_penalty":0.58}}},"rank_prev":28,"rank_delta":0,"image":"https://opengraph.githubassets.com/1/google/langextract","rank":28},{"title":"forrestchang/andrej-karpathy-skills — A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.","url":"https://github.com/forrestchang/andrej-karpathy-skills","source":"github.com","categories":["repo"],"publishedHoursAgo":22,"updatedHoursAgo":22,"createdHoursAgo":1908,"repoAgeDays":79.5,"stars":48186,"forks":3949,"watchers":275,"openIssues":63,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":4,"recentCommits30d":4,"tags":["other"],"score":12.382,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.818,"momentum":10.648,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.15,"buzz":0.0,"hype":0.25,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.818,"velocity":10.648,"score_breakdown":{"trust":0.765,"freshness":0.818,"momentum":10.648,"novelty":0.0,"confirmation":0.0,"spike":0.15,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.536,"recent_stars":3.825,"recent_star_ratio":0.664,"recent_commits":2.173,"network":4.128,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":31,"rank_delta":2,"image":"https://opengraph.githubassets.com/1/forrestchang/andrej-karpathy-skills","rank":29},{"title":"shareAI-lab/learn-claude-code — Bash is all you need -  A nano claude code–like 「agent harness」, built from 0 to 1","url":"https://github.com/shareAI-lab/learn-claude-code","source":"github.com","categories":["repo"],"publishedHoursAgo":47,"updatedHoursAgo":47,"createdHoursAgo":6984,"repoAgeDays":291.0,"stars":54060,"forks":8899,"watchers":248,"openIssues":91,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":2,"recentCommits30d":49,"tags":["AI"],"score":12.318,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.506,"momentum":11.047,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.0,"buzz":0.0,"hype":0.0,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.506,"velocity":11.047,"score_breakdown":{"trust":0.765,"freshness":0.506,"momentum":11.047,"novelty":0.0,"confirmation":0.0,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.226,"recent_stars":3.783,"recent_star_ratio":0.592,"recent_commits":2.749,"network":4.302,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":29,"rank_delta":-1,"image":"https://opengraph.githubassets.com/1/shareAI-lab/learn-claude-code","rank":30},{"title":"datawhalechina/hello-agents — 📚 《从零开始构建智能体》——从零开始的智能体原理与实践教程","url":"https://github.com/datawhalechina/hello-agents","source":"github.com","categories":["repo"],"publishedHoursAgo":35,"updatedHoursAgo":35,"createdHoursAgo":5310,"repoAgeDays":221.2,"stars":37521,"forks":4446,"watchers":122,"openIssues":97,"defaultBranch":"main","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":5,"recentCommits30d":14,"tags":["other"],"score":12.02,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.637,"momentum":11.068,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.0,"buzz":0.0,"hype":0.0,"cross_bonus":0.0,"zh_penalty":0.45,"source":0.9,"fresh":0.637,"velocity":11.068,"score_breakdown":{"trust":0.765,"freshness":0.637,"momentum":11.068,"novelty":0.0,"confirmation":0.0,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.45,"repo":{"recent_push":0.342,"recent_stars":3.934,"recent_star_ratio":0.853,"recent_commits":2.831,"network":3.978,"confidence":0.998,"truncation_penalty":0.58}}},"rank_prev":22,"rank_delta":-9,"image":"https://opengraph.githubassets.com/1/datawhalechina/hello-agents","rank":31},{"title":"sansan0/TrendRadar — ⭐AI-driven public opinion & trend monitor with multi-platform aggregation, RSS, and smart alerts.🎯 告别信息过载，你的 AI 舆情监控助手与热点筛选工具！聚合多平台热点 +  RSS 订阅，支持关键词精准筛选。AI 智能筛选新闻 + AI 翻译 +  AI 分析简报直推手机，也支持接入 MCP 架构，赋能 AI 自然语言对话分析、情感洞察与趋势预测等。支持 Docker ，数据本地/云端自持。集成微信/飞书/钉钉/Telegram/邮件/ntfy/bark/slack 等渠道智能推送。","url":"https://github.com/sansan0/TrendRadar","source":"github.com","categories":["repo"],"publishedHoursAgo":110,"updatedHoursAgo":110,"createdHoursAgo":8476,"repoAgeDays":353.2,"stars":51739,"forks":23154,"watchers":192,"openIssues":45,"defaultBranch":"master","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":487,"recentStars30d":800,"recentStarsPerDay7d":69.57,"stargazerHistoryTruncated":true,"recentCommits7d":1,"recentCommits30d":17,"tags":["other"],"score":10.343,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.15,"momentum":9.878,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.0,"buzz":0.0,"hype":0.0,"cross_bonus":0.0,"zh_penalty":0.45,"source":0.9,"fresh":0.15,"velocity":9.878,"score_breakdown":{"trust":0.765,"freshness":0.15,"momentum":9.878,"novelty":0.0,"confirmation":0.0,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.45,"repo":{"recent_push":0.025,"recent_stars":3.479,"recent_star_ratio":0.425,"recent_commits":1.924,"network":4.461,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":30,"rank_delta":-2,"image":"https://opengraph.githubassets.com/1/sansan0/TrendRadar","rank":32},{"title":"sickn33/antigravity-awesome-skills — Installable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.","url":"https://github.com/sickn33/antigravity-awesome-skills","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":2206,"repoAgeDays":91.9,"stars":33424,"forks":5542,"watchers":33424,"openIssues":0,"defaultBranch":"main","citations":0,"starsPerDay":363.58,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":9.695,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":6.121,"novelty":1.025,"confirmation":0.227,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.42,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":6.121,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":6.121,"novelty":1.025,"confirmation":0.227,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":5.022,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":39,"rank_delta":6,"image":"https://opengraph.githubassets.com/1/sickn33/antigravity-awesome-skills","rank":33},{"title":"Gitlawb/openclaude — Open Claude Is Open-source coding-agent CLI for OpenAI, Gemini, DeepSeek, Ollama, Codex, GitHub Models, and 200+ models via OpenAI-compatible APIs.","url":"https://github.com/Gitlawb/openclaude","source":"github.com","categories":["repo"],"publishedHoursAgo":2,"updatedHoursAgo":2,"createdHoursAgo":372,"repoAgeDays":15.5,"stars":21955,"forks":7507,"watchers":21955,"openIssues":133,"defaultBranch":"main","citations":0,"starsPerDay":1413.07,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":9.42,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.203,"momentum":6.019,"novelty":0.935,"confirmation":0.168,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":2.13,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.203,"velocity":6.019,"score_breakdown":{"trust":0.765,"freshness":1.203,"momentum":6.019,"novelty":0.935,"confirmation":0.168,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.073,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.962,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":37,"rank_delta":3,"image":"https://opengraph.githubassets.com/1/Gitlawb/openclaude","rank":34},{"title":"langchain-ai/deepagents — Agent harness built with LangChain and LangGraph. Equipped with a planning tool, a filesystem backend, and the ability to spawn subagents - well-equipped to handle complex agentic tasks.","url":"https://github.com/langchain-ai/deepagents","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":6305,"repoAgeDays":262.7,"stars":20960,"forks":2913,"watchers":20960,"openIssues":223,"defaultBranch":"main","citations":0,"starsPerDay":79.78,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":9.369,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.25,"momentum":5.873,"novelty":1.025,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.13,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.873,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.873,"novelty":1.025,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.74,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":34,"rank_delta":-1,"image":"https://opengraph.githubassets.com/1/langchain-ai/deepagents","rank":35},{"title":"promptfoo/promptfoo — Test your prompts, agents, and RAGs. Red teaming/pentesting/vulnerability scanning for AI. Compare performance of GPT, Claude, Gemini, Llama, and more. Simple declarative configs with command line and CI/CD integration.  Used by OpenAI and Anthropic.","url":"https://github.com/promptfoo/promptfoo","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":26016,"repoAgeDays":1084.0,"stars":20172,"forks":1743,"watchers":20172,"openIssues":323,"defaultBranch":"main","citations":0,"starsPerDay":18.61,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":9.346,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":5.749,"novelty":1.025,"confirmation":0.227,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.42,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.749,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.749,"novelty":1.025,"confirmation":0.227,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.616,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/promptfoo/promptfoo","rank":36},{"title":"abhigyanpatwari/GitNexus — GitNexus: The Zero-Server Code Intelligence Engine -       GitNexus is a client-side knowledge graph creator that runs entirely in your browser. Drop in a GitHub repo or ZIP file, and get an interactive knowledge graph wit a built in Graph RAG Agent. Perfect for code exploration","url":"https://github.com/abhigyanpatwari/GitNexus","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":6160,"repoAgeDays":256.7,"stars":27685,"forks":3157,"watchers":27685,"openIssues":286,"defaultBranch":"main","citations":0,"starsPerDay":107.85,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":9.318,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":5.939,"novelty":0.931,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.02,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":5.939,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":5.939,"novelty":0.931,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.841,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":32,"rank_delta":-5,"image":"https://opengraph.githubassets.com/1/abhigyanpatwari/GitNexus","rank":37},{"title":"karpathy/autoresearch — AI agents running research on single-GPU nanochat training automatically","url":"https://github.com/karpathy/autoresearch","source":"github.com","categories":["repo"],"publishedHoursAgo":520,"updatedHoursAgo":520,"createdHoursAgo":978,"repoAgeDays":40.8,"stars":73235,"forks":10671,"watchers":592,"openIssues":181,"defaultBranch":"master","citations":0,"starsPerDay":26.67,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"recentStars7d":800,"recentStars30d":800,"recentStarsPerDay7d":114.29,"stargazerHistoryTruncated":true,"recentCommits7d":0,"recentCommits30d":3,"tags":["AI"],"score":9.259,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":0.0,"momentum":8.844,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.0,"buzz":0.0,"hype":0.0,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":0.0,"velocity":8.844,"score_breakdown":{"trust":0.765,"freshness":0.0,"momentum":8.844,"novelty":0.0,"confirmation":0.0,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":0.0,"recent_stars":3.693,"recent_star_ratio":0.437,"recent_commits":0.624,"network":4.534,"confidence":0.999,"truncation_penalty":0.58}}},"rank_prev":33,"rank_delta":-5,"image":"https://opengraph.githubassets.com/1/karpathy/autoresearch","rank":38},{"title":"charmbracelet/crush — Glamourous agentic coding for all 💘","url":"https://github.com/charmbracelet/crush","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":7923,"repoAgeDays":330.2,"stars":23084,"forks":1523,"watchers":23084,"openIssues":382,"defaultBranch":"main","citations":0,"starsPerDay":69.92,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":9.258,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.25,"momentum":5.762,"novelty":1.025,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.13,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.762,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.762,"novelty":1.025,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.626,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":38,"rank_delta":-1,"image":"https://opengraph.githubassets.com/1/charmbracelet/crush","rank":39},{"title":"iOfficeAI/AionUi — Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":6053,"repoAgeDays":252.2,"stars":22005,"forks":1832,"watchers":22005,"openIssues":328,"defaultBranch":"main","citations":0,"starsPerDay":87.24,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["other"],"score":9.193,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.25,"momentum":5.787,"novelty":0.935,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":2.01,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.787,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.787,"novelty":0.935,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.653,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":42,"rank_delta":1,"image":"https://opengraph.githubassets.com/1/iOfficeAI/AionUi","rank":41},{"title":"sgl-project/sglang — SGLang is a high-performance serving framework for large language models and multimodal models.","url":"https://github.com/sgl-project/sglang","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":19907,"repoAgeDays":829.5,"stars":25911,"forks":5401,"watchers":25911,"openIssues":2852,"defaultBranch":"main","citations":0,"starsPerDay":31.24,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":9.198,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":6.075,"novelty":0.694,"confirmation":0.084,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":1.62,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":6.075,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":6.075,"novelty":0.694,"confirmation":0.084,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.94,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":40,"rank_delta":0,"image":"https://opengraph.githubassets.com/1/sgl-project/sglang","rank":40},{"title":"qwibitai/nanoclaw — A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs directly on Anthropic's Agents SDK","url":"https://github.com/qwibitai/nanoclaw","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":1800,"repoAgeDays":75.0,"stars":27378,"forks":12112,"watchers":27378,"openIssues":668,"defaultBranch":"main","citations":0,"starsPerDay":364.97,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":9.12,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.25,"momentum":6.27,"novelty":0.485,"confirmation":0.021,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":1.18,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":6.27,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":6.27,"novelty":0.485,"confirmation":0.021,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":5.134,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":41,"rank_delta":-1,"image":"https://opengraph.githubassets.com/1/qwibitai/nanoclaw","rank":42},{"title":"pipecat-ai/pipecat — Open Source framework for voice and multimodal conversational AI","url":"https://github.com/pipecat-ai/pipecat","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":20187,"repoAgeDays":841.1,"stars":11354,"forks":1940,"watchers":11354,"openIssues":153,"defaultBranch":"main","citations":0,"starsPerDay":13.5,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":9.083,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":5.587,"novelty":1.025,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.13,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.587,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.587,"novelty":1.025,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.467,"confidence":0.995,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/pipecat-ai/pipecat","rank":43},{"title":"ZhuLinsen/daily_stock_analysis — LLM驱动的 A/H/美股智能分析器：多数据源行情 + 实时新闻 + LLM决策仪表盘 + 多渠道推送，零成本定时运行，纯白嫖. LLM-powered stock analysis system for A/H/US markets.","url":"https://github.com/ZhuLinsen/daily_stock_analysis","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":2313,"repoAgeDays":96.4,"stars":30167,"forks":30872,"watchers":30167,"openIssues":70,"defaultBranch":"main","citations":0,"starsPerDay":312.96,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":8.401,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.25,"momentum":6.506,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.45,"source":0.9,"fresh":1.25,"velocity":6.506,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":6.506,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.45,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":5.369,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/ZhuLinsen/daily_stock_analysis","rank":44},{"title":"SpatialEvo: Self-Evolving Spatial Intelligence via Deterministic Geometric Environments","url":"https://huggingface.co/papers/2604.14144","source":"huggingface.co","categories":["paper","hf.daily","self-evolving paradigm","pseudo-labels","3D spatial reasoning","Deterministic Geometric Environment","DGE","point clouds","camera poses","interactive oracles"],"publishedHoursAgo":15,"stars":0,"citations":0,"upvotes":60,"starsPerDay":0,"summary":"SpatialEvo is a self-evolving framework for 3D spatial reasoning that uses deterministic geometric environments to provide objective feedback, enabling efficient training without relying on model consensus.","ai_keywords":["self-evolving paradigm","pseudo-labels","3D spatial reasoning","Deterministic Geometric Environment","DGE","point clouds","camera poses","interactive oracles","shared-parameter policy","task-adaptive scheduler","dynamic curriculum"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.14144.png","paperId":"2604.14144","githubRepo":"https://github.com/ZJU-REAL/SpatialEvo","tags":["other"],"score":3.92,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.909,"momentum":0.96,"novelty":0.836,"confirmation":0.12,"ran_strength":0.0,"spike":0.087,"buzz":0.12,"hype":1.77,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.909,"velocity":0.96,"score_breakdown":{"trust":0.807,"freshness":0.909,"momentum":0.96,"novelty":0.836,"confirmation":0.12,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.14144.png","rank":164},{"title":"CopilotKit/CopilotKit — The Frontend Stack for Agents & Generative UI. React + Angular.  Makers of the AG-UI Protocol","url":"https://github.com/CopilotKit/CopilotKit","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":24780,"repoAgeDays":1032.5,"stars":30266,"forks":3911,"watchers":30266,"openIssues":427,"defaultBranch":"main","citations":0,"starsPerDay":29.31,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["other"],"score":8.398,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":6.053,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":6.053,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":6.053,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.915,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/CopilotKit/CopilotKit","rank":45},{"title":"RationalRewards: Reasoning Rewards Scale Visual Generation Both Training and Test Time","url":"https://huggingface.co/papers/2604.11626","source":"huggingface.co","categories":["paper","hf.daily","reward models","visual generation","preference prediction","reinforcement learning","Generate-Critique-Refine loop","Preference-Anchored Rationalization","PARROT","rationalization"],"publishedHoursAgo":14,"stars":0,"citations":0,"upvotes":92,"starsPerDay":0,"summary":"Training reward models to generate multi-dimensional critiques improves visual generation through both enhanced reinforcement learning rewards and test-time refinement loops, achieving state-of-the-art performance with reduced training data requirements.","ai_keywords":["reward models","visual generation","preference prediction","reinforcement learning","Generate-Critique-Refine loop","Preference-Anchored Rationalization","PARROT","rationalization","structured reasoning","text-to-image generation","image-editing generation","RL-based fine-tuning"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.11626.png","paperId":"2604.11626","githubRepo":"https://github.com/TIGER-AI-Lab/RationalRewards","tags":["other"],"score":3.929,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.918,"momentum":0.96,"novelty":0.836,"confirmation":0.12,"ran_strength":0.0,"spike":0.087,"buzz":0.12,"hype":1.77,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.918,"velocity":0.96,"score_breakdown":{"trust":0.807,"freshness":0.918,"momentum":0.96,"novelty":0.836,"confirmation":0.12,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.11626.png","rank":163},{"title":"github/gh-aw — GitHub Agentic Workflows","url":"https://github.com/github/gh-aw","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":5926,"repoAgeDays":246.9,"stars":4314,"forks":352,"watchers":4314,"openIssues":141,"defaultBranch":"main","citations":0,"starsPerDay":17.47,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":8.38,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":4.884,"novelty":1.025,"confirmation":0.126,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":2.13,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":4.884,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":4.884,"novelty":1.025,"confirmation":0.126,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":3.802,"confidence":0.986,"truncation_penalty":1.0}}},"rank_prev":44,"rank_delta":-2,"image":"https://opengraph.githubassets.com/1/github/gh-aw","rank":46},{"title":"SemaClaw: A Step Towards General-Purpose Personal AI Agents through Harness Engineering","url":"https://huggingface.co/papers/2604.11548","source":"huggingface.co","categories":["paper","hf.daily","multi-agent application framework","harness engineering","agent team orchestration","behavioral safety system","context management architecture","agentic wiki skill"],"publishedHoursAgo":10,"stars":0,"citations":0,"upvotes":15,"starsPerDay":0,"summary":"OpenClaw's emergence in 2026 signifies a shift toward scalable personal AI agents requiring robust infrastructure for control and trustworthiness, addressed by SemaClaw's multi-agent framework with novel orchestration, safety, and context management components.","ai_keywords":["multi-agent application framework","harness engineering","agent team orchestration","behavioral safety system","context management architecture","agentic wiki skill"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.11548.png","paperId":"2604.11548","githubRepo":"https://github.com/midea-ai/SemaClaw","tags":["AI"],"score":3.851,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.954,"momentum":0.77,"novelty":0.836,"confirmation":0.144,"ran_strength":0.0,"spike":0.14,"buzz":0.12,"hype":1.98,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.954,"velocity":0.77,"score_breakdown":{"trust":0.807,"freshness":0.954,"momentum":0.77,"novelty":0.836,"confirmation":0.144,"spike":0.14,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.11548.png","rank":168},{"title":"Unity-Technologies/ml-agents — The Unity Machine Learning Agents Toolkit (ML-Agents) is an open-source project that enables games and simulations to serve as environments for training intelligent agents using deep reinforcement learning and imitation learning.","url":"https://github.com/Unity-Technologies/ml-agents","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":75403,"repoAgeDays":3141.8,"stars":19315,"forks":4440,"watchers":19315,"openIssues":49,"defaultBranch":"develop","citations":0,"starsPerDay":6.15,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":8.285,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":5.94,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.94,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.94,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.808,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/Unity-Technologies/ml-agents","rank":47},{"title":"Memory Transfer Learning: How Memories are Transferred Across Domains in Coding Agents","url":"https://huggingface.co/papers/2604.14004","source":"huggingface.co","categories":["paper","hf.daily","Memory Transfer Learning","memory pool","cross-domain transfer","meta-knowledge","abstraction","memory representation","validation routines","negative transfer"],"publishedHoursAgo":12,"stars":0,"citations":0,"upvotes":23,"starsPerDay":0,"summary":"Memory transfer learning enables cross-domain code generation by leveraging unified memory pools, with performance improvements achieved through high-level abstraction rather than low-level code traces.","ai_keywords":["Memory Transfer Learning","memory pool","cross-domain transfer","meta-knowledge","abstraction","memory representation","validation routines","negative transfer"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.14004.png","paperId":"2604.14004","githubRepo":"https://github.com/KangsanKim07/MemoryTransferLearning","tags":["AI"],"score":3.845,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.935,"momentum":0.859,"novelty":0.759,"confirmation":0.144,"ran_strength":0.0,"spike":0.14,"buzz":0.12,"hype":1.87,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.935,"velocity":0.859,"score_breakdown":{"trust":0.807,"freshness":0.935,"momentum":0.859,"novelty":0.759,"confirmation":0.144,"spike":0.14,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.14004.png","rank":169},{"title":"HKUDS/CLI-Anything — \"CLI-Anything: Making ALL Software Agent-Native\" -- CLI-Hub: https://clianything.cc/","url":"https://github.com/HKUDS/CLI-Anything","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":938,"repoAgeDays":39.1,"stars":31038,"forks":2988,"watchers":31038,"openIssues":38,"defaultBranch":"main","citations":0,"starsPerDay":793.92,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["other"],"score":8.284,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":5.963,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":5.963,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":5.963,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.863,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":76,"rank_delta":28,"image":"https://opengraph.githubassets.com/1/HKUDS/CLI-Anything","rank":48},{"title":"From P(y|x) to P(y): Investigating Reinforcement Learning in Pre-train Space","url":"https://huggingface.co/papers/2604.14142","source":"huggingface.co","categories":["paper","hf.daily","reinforcement learning with verifiable rewards","conditional distribution","marginal distribution","pre-train space","reward-driven online updates","gradient alignment","negative sample reinforcement","policy reincarnation"],"publishedHoursAgo":14,"stars":0,"citations":0,"upvotes":22,"starsPerDay":0,"summary":"PreRL applies reward-driven online updates to the marginal distribution in pre-train space, while DSRL uses NSR-PreRL to expand reasoning horizons before standard RL fine-tuning.","ai_keywords":["reinforcement learning with verifiable rewards","conditional distribution","marginal distribution","pre-train space","reward-driven online updates","gradient alignment","negative sample reinforcement","policy reincarnation","reasoning horizon","standard RL"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.14142.png","paperId":"2604.14142","githubRepo":"https://github.com/Trae1ounG/Pretrain_Space_RLVR","tags":["other"],"score":3.818,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.918,"momentum":0.85,"novelty":0.836,"confirmation":0.12,"ran_strength":0.0,"spike":0.087,"buzz":0.12,"hype":1.77,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.918,"velocity":0.85,"score_breakdown":{"trust":0.807,"freshness":0.918,"momentum":0.85,"novelty":0.836,"confirmation":0.12,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.14142.png","rank":172},{"title":"New ways to create personalized images in the Gemini app","url":"https://blog.google/innovation-and-ai/products/gemini-app/personal-intelligence-nano-banana/","source":"Google AI Blog","categories":["news"],"publishedHoursAgo":1,"stars":0,"citations":0,"starsPerDay":0,"tags":["other"],"score":3.406,"why":{"kind":"social","pool":"social","trust":0.33,"freshness":1.395,"momentum":0.0,"novelty":1.1,"confirmation":0.306,"ran_strength":0.0,"spike":0.275,"buzz":0.0,"hype":2.01,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.6,"fresh":1.395,"velocity":0.0,"score_breakdown":{"trust":0.33,"freshness":1.395,"momentum":0.0,"novelty":1.1,"confirmation":0.306,"spike":0.275,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-ai.svg","rank":186},{"title":"jundot/omlx — LLM inference server with continuous batching & SSD caching for Apple Silicon — managed from the macOS menu bar","url":"https://github.com/jundot/omlx","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":1489,"repoAgeDays":62.1,"stars":10356,"forks":890,"watchers":10356,"openIssues":131,"defaultBranch":"main","citations":0,"starsPerDay":166.82,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":8.233,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":5.387,"novelty":0.396,"confirmation":0.105,"ran_strength":0.0,"spike":0.33,"buzz":0.12,"hype":1.33,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.387,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.387,"novelty":0.396,"confirmation":0.105,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.268,"confidence":0.994,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/jundot/omlx","rank":49},{"title":"Sema Code: Decoupling AI Coding Agents into Programmable, Embeddable Infrastructure","url":"https://huggingface.co/papers/2604.11045","source":"huggingface.co","categories":["paper","hf.daily","AI coding agents","embeddable framework","pluggable architecture","multi-tenant engine isolation","FIFO input queuing","adaptive context compression","multi-agent collaborative scheduling","intelligent Todo-based process management"],"publishedHoursAgo":12,"stars":0,"citations":0,"upvotes":17,"starsPerDay":0,"summary":"Sema Code presents an open AI coding framework that decouples the core agent engine from client interfaces, enabling shared reasoning capabilities across diverse development environments through a standalone npm library and modular architecture.","ai_keywords":["AI coding agents","embeddable framework","pluggable architecture","multi-tenant engine isolation","FIFO input queuing","adaptive context compression","multi-agent collaborative scheduling","intelligent Todo-based process management","asynchronous permission control","ecosystem integration","MCP","Skills","Plugins","background task framework"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.11045.png","paperId":"2604.11045","githubRepo":"https://github.com/midea-ai/sema-code-core","tags":["AI"],"score":3.835,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.935,"momentum":0.796,"novelty":0.836,"confirmation":0.12,"ran_strength":0.0,"spike":0.14,"buzz":0.12,"hype":1.92,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.935,"velocity":0.796,"score_breakdown":{"trust":0.807,"freshness":0.935,"momentum":0.796,"novelty":0.836,"confirmation":0.12,"spike":0.14,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.11045.png","rank":170},{"title":"vercel-labs/agent-browser — Browser automation CLI for AI agents","url":"https://github.com/vercel-labs/agent-browser","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":2290,"repoAgeDays":95.4,"stars":29450,"forks":1787,"watchers":29450,"openIssues":398,"defaultBranch":"main","citations":0,"starsPerDay":308.58,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":8.218,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.25,"momentum":5.873,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.873,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.873,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.735,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/vercel-labs/agent-browser","rank":50},{"title":"Exploration and Exploitation Errors Are Measurable for Language Model Agents","url":"https://huggingface.co/papers/2604.13151","source":"huggingface.co","categories":["paper","hf.daily","language model agents","embodied AI","exploration-exploitation tradeoff","partially observable 2D grid map","Directed Acyclic Graph","policy-agnostic evaluation","reasoning models","harness engineering"],"publishedHoursAgo":13,"stars":0,"citations":0,"upvotes":16,"starsPerDay":0,"summary":"Controllable environments with programmable exploration-exploitation balance are designed to evaluate language model agents' performance on embodied AI tasks, revealing distinct failure modes and demonstrating that reasoning models outperform other approaches.","ai_keywords":["language model agents","embodied AI","exploration-exploitation tradeoff","partially observable 2D grid map","Directed Acyclic Graph","policy-agnostic evaluation","reasoning models","harness engineering"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.13151.png","paperId":"2604.13151","githubRepo":"https://github.com/jjj-madison/measurable-explore-exploit","tags":["AI"],"score":3.761,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.926,"momentum":0.783,"novelty":0.836,"confirmation":0.12,"ran_strength":0.0,"spike":0.087,"buzz":0.12,"hype":1.77,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.926,"velocity":0.783,"score_breakdown":{"trust":0.807,"freshness":0.926,"momentum":0.783,"novelty":0.836,"confirmation":0.12,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.13151.png","rank":173},{"title":"Yeachan-Heo/oh-my-codex — OmX - Oh My codeX: Your codex is not alone. Add hooks, agent teams, HUDs, and so much more.","url":"https://github.com/Yeachan-Heo/oh-my-codex","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":1753,"repoAgeDays":73.1,"stars":23527,"forks":2013,"watchers":23527,"openIssues":7,"defaultBranch":"main","citations":0,"starsPerDay":321.96,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["other"],"score":8.174,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.25,"momentum":5.829,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.829,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.829,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.694,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":72,"rank_delta":20,"image":"https://opengraph.githubassets.com/1/Yeachan-Heo/oh-my-codex","rank":52},{"title":"Correct Prediction, Wrong Steps? Consensus Reasoning Knowledge Graph for Robust Chain-of-Thought Synthesis","url":"http://arxiv.org/abs/2604.14121v1","source":"arxiv.org","categories":["cs.CL"],"publishedHoursAgo":22,"stars":0,"citations":0,"starsPerDay":0,"tags":["other"],"score":3.043,"why":{"kind":"paper","pool":"paper","trust":0.95,"freshness":0.85,"momentum":0.0,"novelty":0.836,"confirmation":0.12,"ran_strength":0.0,"spike":0.087,"buzz":0.12,"hype":1.77,"cross_bonus":0.0,"zh_penalty":0.0,"source":1.0,"fresh":0.85,"velocity":0.0,"score_breakdown":{"trust":0.95,"freshness":0.85,"momentum":0.0,"novelty":0.836,"confirmation":0.12,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-paper.svg","rank":208},{"title":"Gemini Robotics-ER 1.6: 향상된 체화 추론","url":"https://news.hada.io/topic?id=28585","source":"geeknews","categories":["news"],"publishedHoursAgo":13,"stars":0,"citations":0,"starsPerDay":0,"tags":["social"],"score":2.973,"why":{"kind":"social","pool":"social","trust":0.563,"freshness":0.879,"momentum":0.0,"novelty":1.1,"confirmation":0.306,"ran_strength":0.0,"spike":0.125,"buzz":0.0,"hype":1.71,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.66,"fresh":0.879,"velocity":0.0,"score_breakdown":{"trust":0.563,"freshness":0.879,"momentum":0.0,"novelty":1.1,"confirmation":0.306,"spike":0.125,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-ai.svg","rank":219},{"title":"Do Thought Streams Matter? Evaluating Reasoning in Gemini Vision-Language Models for Video Scene Understanding","url":"https://huggingface.co/papers/2604.11177","source":"huggingface.co","categories":["paper","hf.daily","thought streams","vision-language models","video scene understanding","Gemini 2.5 Flash","Flash Lite","contentfulness","thought-final coverage","dominant entity analysis"],"publishedHoursAgo":24,"stars":0,"citations":0,"upvotes":6,"starsPerDay":0,"summary":"Research examines how internal reasoning traces affect video scene understanding in vision-language models, revealing that quality improvements from extended reasoning plateau quickly and that different model variants produce distinct reasoning patterns.","ai_keywords":["thought streams","vision-language models","video scene understanding","Gemini 2.5 Flash","Flash Lite","contentfulness","thought-final coverage","dominant entity analysis","compression-step hallucination"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.11177.png","paperId":"2604.11177","githubRepo":"https://github.com/video-db/gemini-reasoning-eval","tags":["AI"],"score":3.678,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.833,"momentum":0.588,"novelty":0.902,"confirmation":0.26,"ran_strength":0.0,"spike":0.087,"buzz":0.24,"hype":2.24,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.833,"velocity":0.588,"score_breakdown":{"trust":0.807,"freshness":0.833,"momentum":0.588,"novelty":0.902,"confirmation":0.26,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.11177.png","rank":176},{"title":"virattt/dexter — An autonomous agent for deep financial research","url":"https://github.com/virattt/dexter","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":4411,"repoAgeDays":183.8,"stars":21321,"forks":2575,"watchers":21321,"openIssues":71,"defaultBranch":"main","citations":0,"starsPerDay":116.0,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["other"],"score":8.197,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.25,"momentum":5.852,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.852,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.852,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.718,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/virattt/dexter","rank":51},{"title":"Leveraging LLM-GNN Integration for Open-World Question Answering over Knowledge Graphs","url":"http://arxiv.org/abs/2604.13979v1","source":"arxiv.org","categories":["cs.CL","cs.AI","cs.DB"],"publishedHoursAgo":24,"stars":0,"citations":0,"starsPerDay":0,"tags":["AI"],"score":2.974,"why":{"kind":"paper","pool":"paper","trust":0.95,"freshness":0.833,"momentum":0.0,"novelty":0.759,"confirmation":0.144,"ran_strength":0.0,"spike":0.087,"buzz":0.12,"hype":1.72,"cross_bonus":0.0,"zh_penalty":0.0,"source":1.0,"fresh":0.833,"velocity":0.0,"score_breakdown":{"trust":0.95,"freshness":0.833,"momentum":0.0,"novelty":0.759,"confirmation":0.144,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-paper.svg","rank":218},{"title":"macOS용 Gemini 앱 출시","url":"https://news.hada.io/topic?id=28580","source":"geeknews","categories":["news"],"publishedHoursAgo":14,"stars":0,"citations":0,"starsPerDay":0,"tags":["social"],"score":2.94,"why":{"kind":"social","pool":"social","trust":0.563,"freshness":0.846,"momentum":0.0,"novelty":1.1,"confirmation":0.306,"ran_strength":0.0,"spike":0.125,"buzz":0.0,"hype":1.71,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.66,"fresh":0.846,"velocity":0.0,"score_breakdown":{"trust":0.563,"freshness":0.846,"momentum":0.0,"novelty":1.1,"confirmation":0.306,"spike":0.125,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-ai.svg","rank":227},{"title":"mastra-ai/mastra — From the team behind Gatsby, Mastra is a framework for building AI-powered applications and agents with a modern TypeScript stack.","url":"https://github.com/mastra-ai/mastra","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":14827,"repoAgeDays":617.8,"stars":23059,"forks":1901,"watchers":23059,"openIssues":368,"defaultBranch":"main","citations":0,"starsPerDay":37.32,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":8.155,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":5.81,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.81,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.81,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.675,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/mastra-ai/mastra","rank":53},{"title":"KnowRL: Boosting LLM Reasoning via Reinforcement Learning with Minimal-Sufficient Knowledge Guidance","url":"https://huggingface.co/papers/2604.12627","source":"huggingface.co","categories":["paper","hf.daily","reinforcement learning","reasoning","reward sparsity","hint-based RL","minimal-sufficient guidance","atomic knowledge points","constrained subset search","pruning interaction paradox"],"publishedHoursAgo":37,"stars":0,"citations":0,"upvotes":92,"starsPerDay":0,"summary":"KnowRL is a knowledge-guided reinforcement learning framework that improves reasoning in language models by optimizing compact, interaction-aware guidance subsets through constrained subset search and addressing pruning interaction paradoxes.","ai_keywords":["reinforcement learning","reasoning","reward sparsity","hint-based RL","minimal-sufficient guidance","atomic knowledge points","constrained subset search","pruning interaction paradox","subset curation","language models"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.12627.png","paperId":"2604.12627","githubRepo":"https://github.com/Hasuer/KnowRL","tags":["AI"],"score":3.659,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.735,"momentum":0.96,"novelty":0.836,"confirmation":0.12,"ran_strength":0.0,"spike":0.0,"buzz":0.12,"hype":1.52,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.735,"velocity":0.96,"score_breakdown":{"trust":0.807,"freshness":0.735,"momentum":0.96,"novelty":0.836,"confirmation":0.12,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":54,"rank_delta":-123,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.12627.png","rank":177},{"title":"MERRIN: A Benchmark for Multimodal Evidence Retrieval and Reasoning in Noisy Web Environments","url":"https://huggingface.co/papers/2604.13418","source":"huggingface.co","categories":["paper","hf.daily","search-augmented agents","multimodal evidence retrieval","multi-hop reasoning","noisy web environments","human-annotated benchmark","web search","multimodal evidence","search agents"],"publishedHoursAgo":14,"stars":0,"citations":0,"upvotes":4,"starsPerDay":0,"summary":"MERRIN is a human-annotated benchmark for evaluating search-augmented agents in multimodal, noisy web environments, demonstrating significant challenges in retrieving and reasoning over diverse evidence types.","ai_keywords":["search-augmented agents","multimodal evidence retrieval","multi-hop reasoning","noisy web environments","human-annotated benchmark","web search","multimodal evidence","search agents","artificial intelligence agents","web retrieval"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.13418.png","paperId":"2604.13418","githubRepo":"https://github.com/HanNight/MERRIN","tags":["AI"],"score":3.645,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.918,"momentum":0.514,"novelty":0.902,"confirmation":0.216,"ran_strength":0.0,"spike":0.087,"buzz":0.24,"hype":2.13,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.918,"velocity":0.514,"score_breakdown":{"trust":0.807,"freshness":0.918,"momentum":0.514,"novelty":0.902,"confirmation":0.216,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.13418.png","rank":178},{"title":"ToolOmni: Enabling Open-World Tool Use via Agentic learning with Proactive Retrieval and Grounded Execution","url":"http://arxiv.org/abs/2604.13787v1","source":"arxiv.org","categories":["cs.CL"],"publishedHoursAgo":27,"stars":0,"citations":0,"starsPerDay":0,"tags":["AI"],"score":2.94,"why":{"kind":"paper","pool":"paper","trust":0.95,"freshness":0.81,"momentum":0.0,"novelty":0.836,"confirmation":0.144,"ran_strength":0.0,"spike":0.0,"buzz":0.12,"hype":1.58,"cross_bonus":0.0,"zh_penalty":0.0,"source":1.0,"fresh":0.81,"velocity":0.0,"score_breakdown":{"trust":0.95,"freshness":0.81,"momentum":0.0,"novelty":0.836,"confirmation":0.144,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-paper.svg","rank":226},{"title":"coze-dev/coze-studio — An AI agent development platform with all-in-one visual tools, simplifying agent creation, debugging, and deployment like never before. Coze your way to AI Agent creation.","url":"https://github.com/coze-dev/coze-studio","source":"github.com","categories":["repo"],"publishedHoursAgo":2,"updatedHoursAgo":2,"createdHoursAgo":7069,"repoAgeDays":294.6,"stars":20539,"forks":2979,"watchers":20539,"openIssues":492,"defaultBranch":"main","citations":0,"starsPerDay":69.72,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":8.093,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.203,"momentum":5.795,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.203,"velocity":5.795,"score_breakdown":{"trust":0.765,"freshness":1.203,"momentum":5.795,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.073,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.739,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":50,"rank_delta":-6,"image":"https://opengraph.githubassets.com/1/coze-dev/coze-studio","rank":56},{"title":"The cognitive companion: a lightweight parallel monitoring architecture for detecting and recovering from reasoning degradation in LLM agents","url":"http://arxiv.org/abs/2604.13759v1","source":"arxiv.org","categories":["cs.AI","cs.LG"],"publishedHoursAgo":28,"stars":0,"citations":0,"starsPerDay":0,"tags":["AI"],"score":2.908,"why":{"kind":"paper","pool":"paper","trust":0.95,"freshness":0.802,"momentum":0.0,"novelty":0.836,"confirmation":0.12,"ran_strength":0.0,"spike":0.0,"buzz":0.12,"hype":1.52,"cross_bonus":0.0,"zh_penalty":0.0,"source":1.0,"fresh":0.802,"velocity":0.0,"score_breakdown":{"trust":0.95,"freshness":0.802,"momentum":0.0,"novelty":0.836,"confirmation":0.12,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-paper.svg","rank":230},{"title":"GameWorld: Towards Standardized and Verifiable Evaluation of Multimodal Game Agents","url":"https://huggingface.co/papers/2604.07429","source":"huggingface.co","categories":["paper","hf.daily","Multimodal Large Language Model","video games","browser environments","computer-use agents","semantic action space","semantic action parsing","state-verifiable metrics","action interfaces"],"publishedHoursAgo":15,"stars":0,"citations":0,"upvotes":104,"starsPerDay":0,"summary":"GameWorld presents a standardized benchmark for evaluating multimodal large language model agents in video games, featuring diverse games and verified metrics for comprehensive assessment.","ai_keywords":["Multimodal Large Language Model","video games","browser environments","computer-use agents","semantic action space","semantic action parsing","state-verifiable metrics","action interfaces","game agent interfaces"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.07429.png","paperId":"2604.07429","githubRepo":"https://github.com/gameworld-project/gameworld","tags":["AI"],"score":3.623,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.909,"momentum":0.96,"novelty":0.563,"confirmation":0.096,"ran_strength":0.0,"spike":0.087,"buzz":0.12,"hype":1.32,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.909,"velocity":0.96,"score_breakdown":{"trust":0.807,"freshness":0.909,"momentum":0.96,"novelty":0.563,"confirmation":0.096,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.07429.png","rank":179},{"title":"oraios/serena — A powerful MCP toolkit for coding, providing semantic retrieval and editing capabilities  - the IDE for your agent","url":"https://github.com/oraios/serena","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":9329,"repoAgeDays":388.7,"stars":23029,"forks":1547,"watchers":23029,"openIssues":92,"defaultBranch":"main","citations":0,"starsPerDay":59.24,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":8.109,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":5.764,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.764,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.764,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.629,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/oraios/serena","rank":55},{"title":"rtk-ai/rtk — CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies","url":"https://github.com/rtk-ai/rtk","source":"github.com","categories":["repo"],"publishedHoursAgo":2,"updatedHoursAgo":2,"createdHoursAgo":2015,"repoAgeDays":84.0,"stars":27876,"forks":1636,"watchers":27876,"openIssues":597,"defaultBranch":"master","citations":0,"starsPerDay":331.98,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":8.057,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.203,"momentum":5.759,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.203,"velocity":5.759,"score_breakdown":{"trust":0.765,"freshness":1.203,"momentum":5.759,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.073,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.699,"confidence":0.998,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/rtk-ai/rtk","rank":58},{"title":"Character Beyond Speech: Leveraging Role-Playing Evaluation in Audio Large Language Models via Reinforcement Learning","url":"http://arxiv.org/abs/2604.13804v1","source":"arxiv.org","categories":["cs.LG"],"publishedHoursAgo":27,"stars":0,"citations":0,"starsPerDay":0,"tags":["AI"],"score":2.863,"why":{"kind":"paper","pool":"paper","trust":0.95,"freshness":0.81,"momentum":0.0,"novelty":0.759,"confirmation":0.144,"ran_strength":0.0,"spike":0.0,"buzz":0.12,"hype":1.47,"cross_bonus":0.0,"zh_penalty":0.0,"source":1.0,"fresh":0.81,"velocity":0.0,"score_breakdown":{"trust":0.95,"freshness":0.81,"momentum":0.0,"novelty":0.759,"confirmation":0.144,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-paper.svg","rank":234},{"title":"⚡ Meet Qwen3.6-35B-A3B：Now Open-Source！🚀🚀 A sparse MoE model, 35B total params, 3B active. Apache 2.0 license. 🔥 Agentic coding on par with models 10x its active size 📷 Strong multimodal perception and reasoning ability 🧠 Multimodal thinking + n - x.com","url":"https://news.google.com/rss/articles/CBMiY0FVX3lxTE5kaTlvYXg5WGlmWlM4bzF4Tkt0UTVQWTlRb0xXMU92TmNzMFBETkQyRE40eFdYUkJJWG8yM0tQUlhQZ0g0bThkNmtITFNlTmlOR2g5MmtXckhTLW03TWEyRDlySQ?oc=5","source":"x.com","categories":["news"],"publishedHoursAgo":72,"stars":0,"citations":0,"starsPerDay":0,"tags":["AI","social"],"score":2.358,"why":{"kind":"social","pool":"social","trust":0.399,"freshness":0.091,"momentum":0.0,"novelty":1.316,"confirmation":0.552,"ran_strength":0.0,"spike":0.0,"buzz":0.48,"hype":2.23,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.58,"fresh":0.091,"velocity":0.0,"score_breakdown":{"trust":0.399,"freshness":0.091,"momentum":0.0,"novelty":1.316,"confirmation":0.552,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-ai.svg","rank":253},{"title":"Many-Tier Instruction Hierarchy in LLM Agents","url":"https://huggingface.co/papers/2604.09443","source":"huggingface.co","categories":["paper","hf.daily","instruction hierarchy","large language model agents","instruction conflict resolution","privilege levels","Many-Tier Instruction Hierarchy","ManyIH-Bench","agentic tasks","instruction-following"],"publishedHoursAgo":37,"stars":0,"citations":0,"upvotes":14,"starsPerDay":0,"summary":"Large language model agents require robust instruction conflict resolution mechanisms that can handle arbitrary privilege levels across diverse real-world scenarios, revealing current models' limitations in managing complex hierarchical instructions.","ai_keywords":["instruction hierarchy","large language model agents","instruction conflict resolution","privilege levels","Many-Tier Instruction Hierarchy","ManyIH-Bench","agentic tasks","instruction-following","coding tasks"],"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.09443.png","paperId":"2604.09443","githubRepo":"https://github.com/JHU-CLSP/ManyIH","tags":["AI"],"score":3.479,"why":{"kind":"paper","pool":"paper","trust":0.807,"freshness":0.735,"momentum":0.756,"novelty":0.836,"confirmation":0.144,"ran_strength":0.0,"spike":0.0,"buzz":0.12,"hype":1.58,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.85,"fresh":0.735,"velocity":0.756,"score_breakdown":{"trust":0.807,"freshness":0.735,"momentum":0.756,"novelty":0.836,"confirmation":0.144,"spike":0.0,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":48,"rank_delta":-135,"image":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2604.09443.png","rank":183},{"title":"QwenLM/qwen-code — An open-source AI agent that lives in your terminal.","url":"https://github.com/QwenLM/qwen-code","source":"github.com","categories":["repo"],"publishedHoursAgo":1,"updatedHoursAgo":1,"createdHoursAgo":7070,"repoAgeDays":294.6,"stars":23370,"forks":2202,"watchers":23370,"openIssues":768,"defaultBranch":"main","citations":0,"starsPerDay":79.33,"ghPool":"breakout","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.1,"tags":["AI"],"score":8.128,"why":{"kind":"repo","pool":"breakout","trust":0.765,"freshness":1.226,"momentum":5.807,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.226,"velocity":5.807,"score_breakdown":{"trust":0.765,"freshness":1.226,"momentum":5.807,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.111,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.711,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":55,"rank_delta":1,"image":"https://opengraph.githubassets.com/1/QwenLM/qwen-code","rank":54},{"title":"Decoding the Delta: Unifying Remote Sensing Change Detection and Understanding with Multimodal Large Language Models","url":"http://arxiv.org/abs/2604.14044v1","source":"arxiv.org","categories":["cs.CV"],"publishedHoursAgo":23,"stars":0,"citations":0,"starsPerDay":0,"tags":["AI"],"score":2.738,"why":{"kind":"paper","pool":"paper","trust":0.95,"freshness":0.841,"momentum":0.0,"novelty":0.563,"confirmation":0.096,"ran_strength":0.0,"spike":0.087,"buzz":0.12,"hype":1.32,"cross_bonus":0.0,"zh_penalty":0.0,"source":1.0,"fresh":0.841,"velocity":0.0,"score_breakdown":{"trust":0.95,"freshness":0.841,"momentum":0.0,"novelty":0.563,"confirmation":0.096,"spike":0.087,"cross_bonus":0.0,"zh_penalty":0.0}},"rank_prev":null,"rank_delta":null,"image":"/assets/thumb-paper.svg","rank":239},{"title":"screenpipe/screenpipe — Run agents that work for you based on what you do. AI finally knows what you are doing","url":"https://github.com/screenpipe/screenpipe","source":"github.com","categories":["repo"],"publishedHoursAgo":0,"updatedHoursAgo":0,"createdHoursAgo":15986,"repoAgeDays":666.1,"stars":18225,"forks":1621,"watchers":18225,"openIssues":38,"defaultBranch":"main","citations":0,"starsPerDay":27.36,"ghPool":"maintained","ghQuery":"agentic OR llm OR rag OR multimodal","poolPriority":1.0,"tags":["AI"],"score":8.046,"why":{"kind":"repo","pool":"maintained","trust":0.765,"freshness":1.25,"momentum":5.701,"novelty":0.0,"confirmation":0.0,"ran_strength":0.0,"spike":0.33,"buzz":0.0,"hype":0.55,"cross_bonus":0.0,"zh_penalty":0.0,"source":0.9,"fresh":1.25,"velocity":5.701,"score_breakdown":{"trust":0.765,"freshness":1.25,"momentum":5.701,"novelty":0.0,"confirmation":0.0,"spike":0.33,"cross_bonus":0.0,"zh_penalty":0.0,"repo":{"recent_push":1.15,"recent_stars":0.0,"recent_star_ratio":0.0,"recent_commits":0.0,"network":4.569,"confidence":0.997,"truncation_penalty":1.0}}},"rank_prev":null,"rank_delta":null,"image":"https://opengraph.githubassets.com/1/screenpipe/screenpipe","rank":60}],"sourceCounts":{"github.com":58,"huggingface.co":12,"Google AI Blog":1,"arxiv.org":6,"geeknews":2,"x.com":1},"paperCount":18}