Saiten MCP Server

# 🏆 Agents League @ TechConnect — Scoring Ranking > Auto-generated: 2026-02-13 13:51:50 UTC > Scored: 52 / 0 submissions --- ## 🥇 Top 10 | Rank | Project | Track | Submitter | Score | |------|---------|-------|-----------|-------| | 1 | [TrendSurf](https://github.com/microsoft/agentsleague-techconnect/issues/27) | 🧠 | [@ethanjenkins1](https://github.com/ethanjenkins1) (hedinulo) | 91.5 | | 2 | [Zavatravel](https://github.com/microsoft/agentsleague-techconnect/issues/38) | 🧠 | [@sureshpaulraj](https://github.com/sureshpaulraj) (surep, senthkumar, makizhne) | 91.5 | | 3 | [Social AI Studio](https://github.com/microsoft/agentsleague-techconnect/issues/61) | 🧠 | [@naoki1213mj](https://github.com/naoki1213mj) | 91.5 | | 4 | [HeadlineArt: Where News Meets Canvas](https://github.com/microsoft/agentsleague-techconnect/issues/52) | 🧠 | [@FranGenoa, Ajananth, Lakshmy](https://github.com/FranGenoa, Ajananth, Lakshmy) (frfava, ajananth, lchandran, msadhwani, kumarahu) | 91.5 | | 5 | [Saiten — Multi-Agent Scoring System](https://github.com/microsoft/agentsleague-techconnect/issues/54) | 🎨 | [@aktsmm](https://github.com/aktsmm) (tatsumiy, nmatsumoto, takuyamasuda) | 90.0 | | 6 | [EasyExpenseAI](https://github.com/microsoft/agentsleague-techconnect/issues/49) | 🎨 | [@chakras](https://github.com/chakras) (arashinkar,jaideeproy) | 90.0 | | 7 | [StoryCircuit](https://github.com/microsoft/agentsleague-techconnect/issues/58) | 🧠 | [@vrengar](https://github.com/vrengar) | 88.0 | | 8 | [SovereignFit](https://github.com/microsoft/agentsleague-techconnect/issues/32) | 🎨 | [@danieletten](https://github.com/danieletten) (daetten) | 86.1 | | 9 | [Zava Smart Assistant](https://github.com/microsoft/agentsleague-techconnect/issues/42) | 🎨 | [@payton-chou-ms](https://github.com/payton-chou-ms) | 85.6 | | 10 | [techconnect-policy-compliance-checker](https://github.com/microsoft/agentsleague-techconnect/issues/35) | 🧠 | [@taimoor-janjua](https://github.com/taimoor-janjua) | 85.5 | --- ## 🏅 Track Top 3 ### 🎨 Creative Apps | Rank | Project | Submitter | Score | |------|---------|-----------|-------| | 1 | [Saiten — Multi-Agent Scoring System](https://github.com/microsoft/agentsleague-techconnect/issues/54) | [@aktsmm](https://github.com/aktsmm) (tatsumiy, nmatsumoto, takuyamasuda) | 90.0 | | 2 | [EasyExpenseAI](https://github.com/microsoft/agentsleague-techconnect/issues/49) | [@chakras](https://github.com/chakras) (arashinkar,jaideeproy) | 90.0 | | 3 | [SovereignFit](https://github.com/microsoft/agentsleague-techconnect/issues/32) | [@danieletten](https://github.com/danieletten) (daetten) | 86.1 | ### 🧠 Reasoning Agents | Rank | Project | Submitter | Score | |------|---------|-----------|-------| | 1 | [TrendSurf](https://github.com/microsoft/agentsleague-techconnect/issues/27) | [@ethanjenkins1](https://github.com/ethanjenkins1) (hedinulo) | 91.5 | | 2 | [Zavatravel](https://github.com/microsoft/agentsleague-techconnect/issues/38) | [@sureshpaulraj](https://github.com/sureshpaulraj) (surep, senthkumar, makizhne) | 91.5 | | 3 | [Social AI Studio](https://github.com/microsoft/agentsleague-techconnect/issues/61) | [@naoki1213mj](https://github.com/naoki1213mj) | 91.5 | ### 💼 Enterprise Agents | Rank | Project | Submitter | Score | |------|---------|-----------|-------| | 1 | [SE Agent](https://github.com/microsoft/agentsleague-techconnect/issues/13) | [@Daboostr](https://github.com/Daboostr) | 83.4 | | 2 | [Compliment my pet](https://github.com/microsoft/agentsleague-techconnect/issues/56) | [@tanjaquaijtaal-ux](https://github.com/tanjaquaijtaal-ux) | 60.1 | | 3 | [Corporate Bullshit Translator](https://github.com/microsoft/agentsleague-techconnect/issues/44) | [@Aakanksha346](https://github.com/Aakanksha346) | 56.6 | --- ## 📊 All Submissions | # | Issue | Project | Track | Submitter | Repo | Score | Scored At | |---|-------|---------|-------|-----------|------|-------|-----------| | 1 | [#27](https://github.com/microsoft/agentsleague-techconnect/issues/27) | [TrendSurf](https://github.com/microsoft/agentsleague-techconnect/issues/27) | 🧠 | [@ethanjenkins1](https://github.com/ethanjenkins1) (hedinulo) | [repo](https://github.com/ethanjenkins1/trendsurf) | 91.5 | 2026-02-13 | | 2 | [#38](https://github.com/microsoft/agentsleague-techconnect/issues/38) | [Zavatravel](https://github.com/microsoft/agentsleague-techconnect/issues/38) | 🧠 | [@sureshpaulraj](https://github.com/sureshpaulraj) (surep, senthkumar, makizhne) | [repo](https://github.com/sureshpaulraj/zavatravel) | 91.5 | 2026-02-13 | | 3 | [#61](https://github.com/microsoft/agentsleague-techconnect/issues/61) | [Social AI Studio](https://github.com/microsoft/agentsleague-techconnect/issues/61) | 🧠 | [@naoki1213mj](https://github.com/naoki1213mj) | [repo](https://github.com/naoki1213mj/social-ai-studio) | 91.5 | 2026-02-13 | | 4 | [#52](https://github.com/microsoft/agentsleague-techconnect/issues/52) | [HeadlineArt: Where News Meets Canvas](https://github.com/microsoft/agentsleague-techconnect/issues/52) | 🧠 | [@FranGenoa, Ajananth, Lakshmy](https://github.com/FranGenoa, Ajananth, Lakshmy) (frfava, ajananth, lchandran, msadhwani, kumarahu) | [repo](https://github.com/FranGenoa/HeadlineArt) | 91.5 | 2026-02-13 | | 5 | [#54](https://github.com/microsoft/agentsleague-techconnect/issues/54) | [Saiten — Multi-Agent Scoring System](https://github.com/microsoft/agentsleague-techconnect/issues/54) | 🎨 | [@aktsmm](https://github.com/aktsmm) (tatsumiy, nmatsumoto, takuyamasuda) | [repo](https://github.com/aktsmm/FY26_techconnect_saiten) | 90.0 | 2026-02-13 | | 6 | [#49](https://github.com/microsoft/agentsleague-techconnect/issues/49) | [EasyExpenseAI](https://github.com/microsoft/agentsleague-techconnect/issues/49) | 🎨 | [@chakras](https://github.com/chakras) (arashinkar,jaideeproy) | [repo](https://github.com/chakras/Easy-Expense-AI) | 90.0 | 2026-02-13 | | 7 | [#58](https://github.com/microsoft/agentsleague-techconnect/issues/58) | [StoryCircuit](https://github.com/microsoft/agentsleague-techconnect/issues/58) | 🧠 | [@vrengar](https://github.com/vrengar) | [repo](https://github.com/vrengar/storycircuit) | 88.0 | 2026-02-13 | | 8 | [#32](https://github.com/microsoft/agentsleague-techconnect/issues/32) | [SovereignFit](https://github.com/microsoft/agentsleague-techconnect/issues/32) | 🎨 | [@danieletten](https://github.com/danieletten) (daetten) | [repo](https://github.com/danieletten/SovereignFit) | 86.1 | 2026-02-13 | | 9 | [#42](https://github.com/microsoft/agentsleague-techconnect/issues/42) | [Zava Smart Assistant](https://github.com/microsoft/agentsleague-techconnect/issues/42) | 🎨 | [@payton-chou-ms](https://github.com/payton-chou-ms) | [repo](https://github.com/payton-chou-ms/poc-agents-league-techconnect-2026) | 85.6 | 2026-02-13 | | 10 | [#35](https://github.com/microsoft/agentsleague-techconnect/issues/35) | [techconnect-policy-compliance-checker](https://github.com/microsoft/agentsleague-techconnect/issues/35) | 🧠 | [@taimoor-janjua](https://github.com/taimoor-janjua) | [repo](https://github.com/taimoor-janjua/techconnect-policy-compliance-checker) | 85.5 | 2026-02-13 | | 11 | [#60](https://github.com/microsoft/agentsleague-techconnect/issues/60) | [BrandComm-agent](https://github.com/microsoft/agentsleague-techconnect/issues/60) | 🧠 | [@joyjeet](https://github.com/joyjeet) | [repo](https://github.com/joyjeet/TechConnectHack/tree/main/agent-webapp) | 84.0 | 2026-02-13 | | 12 | [#13](https://github.com/microsoft/agentsleague-techconnect/issues/13) | [SE Agent](https://github.com/microsoft/agentsleague-techconnect/issues/13) | 💼 | [@Daboostr](https://github.com/Daboostr) | [repo](https://github.com/daboostr/SEDemoAgent) | 83.4 | 2026-02-13 | | 13 | [#34](https://github.com/microsoft/agentsleague-techconnect/issues/34) | [LearnIQ](https://github.com/microsoft/agentsleague-techconnect/issues/34) | 🎨 | [@kylednoland](https://github.com/kylednoland) | [repo](https://github.com/kylednoland/LearnIQ) | 82.2 | 2026-02-13 | | 14 | [#15](https://github.com/microsoft/agentsleague-techconnect/issues/15) | [Skills Hub](https://github.com/microsoft/agentsleague-techconnect/issues/15) | 🎨 | [@samueltauil, @rleme](https://github.com/samueltauil, @rleme) (samuel.tauil, rodrigoleme) | [repo](https://github.com/samueltauil/skills-hub) | 81.7 | 2026-02-13 | | 15 | [#31](https://github.com/microsoft/agentsleague-techconnect/issues/31) | [Threat Incident Swarm Commander](https://github.com/microsoft/agentsleague-techconnect/issues/31) | 🎨 | [@dondinulos](https://github.com/dondinulos) (ethanjenkins) | [repo](https://github.com/dondinulos/matrix-smith-swarm) | 81.7 | 2026-02-13 | | 16 | [#47](https://github.com/microsoft/agentsleague-techconnect/issues/47) | [Semantic Kernel and AutoGen to MAF Modernizer](https://github.com/microsoft/agentsleague-techconnect/issues/47) | 🧠 | [@misantibanez, valegrajales, eraggi-ar, afrancoc2000](https://github.com/misantibanez, valegrajales, eraggi-ar, afrancoc2000) (michelle.santibanez, valentinag, emilioraggi, anafranco) | [repo](https://github.com/afrancoc2000/tech-connect-2026-sk-modernizer) | 80.5 | 2026-02-13 | | 17 | [#55](https://github.com/microsoft/agentsleague-techconnect/issues/55) | [kube-copilot](https://github.com/microsoft/agentsleague-techconnect/issues/55) | 🎨 | [@openjny](https://github.com/openjny) | [repo](https://github.com/openjny/kube-copilot) | 80.0 | 2026-02-13 | | 18 | [#12](https://github.com/microsoft/agentsleague-techconnect/issues/12) | [Zava Content Studio — Social Media Content Agent](https://github.com/microsoft/agentsleague-techconnect/issues/12) | 🧠 | [@lordlinus, @ganachan, @hieunhums](https://github.com/lordlinus, @ganachan, @hieunhums) (ssattiraju, ganac, hieunhums) | [repo](https://github.com/lordlinus/hackfest-fy26) | 79.5 | 2026-02-13 | | 19 | [#16](https://github.com/microsoft/agentsleague-techconnect/issues/16) | [FoundryReleaseScout](https://github.com/microsoft/agentsleague-techconnect/issues/16) | 🧠 | [@JW-Sthlm](https://github.com/JW-Sthlm) (jwallquist) | [repo](https://github.com/JW-Sthlm/jw-hackfest/) | 79.0 | 2026-02-13 | | 20 | [#39](https://github.com/microsoft/agentsleague-techconnect/issues/39) | [CodeIntel](https://github.com/microsoft/agentsleague-techconnect/issues/39) | 🎨 | [@Mazher-Ul-Haq](https://github.com/Mazher-Ul-Haq) | [repo](https://github.com/Mazher-Ul-Haq/codeintel) | 77.8 | 2026-02-13 | | 21 | [#18](https://github.com/microsoft/agentsleague-techconnect/issues/18) | [My Proposal Develop](https://github.com/microsoft/agentsleague-techconnect/issues/18) | 🧠 | [@balakreshnan](https://github.com/balakreshnan) | [repo](https://github.com/balakreshnan/TechconnectFY26Hackfest) | 77.0 | 2026-02-13 | | 22 | [#43](https://github.com/microsoft/agentsleague-techconnect/issues/43) | [Nimbus Content Agent](https://github.com/microsoft/agentsleague-techconnect/issues/43) | 🧠 | [@namratasherkhane](https://github.com/namratasherkhane) | [repo](https://github.com/namratasherkhane/namratasherkhane) | 76.5 | 2026-02-13 | | 23 | [#23](https://github.com/microsoft/agentsleague-techconnect/issues/23) | [repo-analyzer](https://github.com/microsoft/agentsleague-techconnect/issues/23) | 🎨 | [@olivomarco](https://github.com/olivomarco) | [repo](https://github.com/olivomarco/repo-analyzer) | 76.1 | 2026-02-13 | | 24 | [#53](https://github.com/microsoft/agentsleague-techconnect/issues/53) | [Marketing Asset Creation Agent](https://github.com/microsoft/agentsleague-techconnect/issues/53) | 🎨 | [@jetzlstorfer @stephan-nik](https://github.com/jetzlstorfer @stephan-nik) (jetzlstorfer, sniklas) | [repo](https://github.com/jetzlstorfer/marketing-asset-creation-agent) | 75.6 | 2026-02-13 | | 25 | [#29](https://github.com/microsoft/agentsleague-techconnect/issues/29) | [Upgrade Sage](https://github.com/microsoft/agentsleague-techconnect/issues/29) | 🧠 | [@betanu701](https://github.com/betanu701) | [repo](https://github.com/Betanu701/UpgradeSage) | 74.5 | 2026-02-13 | | 26 | [#37](https://github.com/microsoft/agentsleague-techconnect/issues/37) | [Making money by poop](https://github.com/microsoft/agentsleague-techconnect/issues/37) | 🎨 | [@bensonwugit](https://github.com/bensonwugit) (benswu, YENJUSU, thhuan, huihuan, yumwang) | [repo](https://github.com/bensonwugit/make_money_by_shit) | 73.9 | 2026-02-13 | | 27 | [#48](https://github.com/microsoft/agentsleague-techconnect/issues/48) | [CommsGuard](https://github.com/microsoft/agentsleague-techconnect/issues/48) | 🧠 | [@Akshay8870](https://github.com/Akshay8870) (sankhac , jaideeproy) | [repo](https://github.com/Akshay8870/bfsi-commsguard.git) | 73.5 | 2026-02-13 | | 28 | [#33](https://github.com/microsoft/agentsleague-techconnect/issues/33) | [BeatMyPrompt](https://github.com/microsoft/agentsleague-techconnect/issues/33) | 🎨 | [@fimdim](https://github.com/fimdim) | [repo](https://github.com/fimdim/BeatMyPrompt) | 72.8 | 2026-02-13 | | 29 | [#22](https://github.com/microsoft/agentsleague-techconnect/issues/22) | [DailySync](https://github.com/microsoft/agentsleague-techconnect/issues/22) | 🎨 | [@robertoborges](https://github.com/robertoborges) (roborges) | [repo](https://github.com/RobertoBorges/techconnect-Hackfest) | 71.1 | 2026-02-13 | | 30 | [#41](https://github.com/microsoft/agentsleague-techconnect/issues/41) | [Architect Agent](https://github.com/microsoft/agentsleague-techconnect/issues/41) | 🎨 | [@bzbrkic](https://github.com/bzbrkic) | [repo](https://github.com/bzbrkic/Agent-Architect) | 69.5 | 2026-02-13 | | 31 | [#40](https://github.com/microsoft/agentsleague-techconnect/issues/40) | [AI Project Recap Storyboard](https://github.com/microsoft/agentsleague-techconnect/issues/40) | 🎨 | [@huier23, dinowang, chhtw, Phonicdream](https://github.com/huier23, dinowang, chhtw, Phonicdream) (huihuan, yumwang, thhuan, yenjusu) | [repo](https://github.com/where-is-my-org/Creative-Apps) | 66.1 | 2026-02-13 | | 32 | [#59](https://github.com/microsoft/agentsleague-techconnect/issues/59) | [Dictation 4 Kids](https://github.com/microsoft/agentsleague-techconnect/issues/59) | 🎨 | [@kevintkliu](https://github.com/kevintkliu) (kevinliu2, erikatang) | [repo](https://github.com/kevintkliu/dictation4kids) | 66.1 | 2026-02-13 | | 33 | [#25](https://github.com/microsoft/agentsleague-techconnect/issues/25) | [Team Risk Analyzer](https://github.com/microsoft/agentsleague-techconnect/issues/25) | 🎨 | [@rebmid](https://github.com/rebmid) | [repo](https://github.com/rebmid/team-risk-analyzer) | 65.6 | 2026-02-13 | | 34 | [#21](https://github.com/microsoft/agentsleague-techconnect/issues/21) | [Deal to Glory](https://github.com/microsoft/agentsleague-techconnect/issues/21) | 🎨 | [@billwhalenmsft](https://github.com/billwhalenmsft) | [repo](https://github.com/billwhalenmsft/agents-league-deal-to-glory**) | 63.9 | 2026-02-13 | | 35 | [#45](https://github.com/microsoft/agentsleague-techconnect/issues/45) | [Prompt Escape - The AI Escape Room Craft the right prompts. Solve the puzzles. Escape.](https://github.com/microsoft/agentsleague-techconnect/issues/45) | 🎨 | [@billwhalenmsft](https://github.com/billwhalenmsft) | [repo](https://github.com/billwhalenmsft/PromptEscape) | 63.9 | 2026-02-13 | | 36 | [#46](https://github.com/microsoft/agentsleague-techconnect/issues/46) | [PowerPoint Reviewer Agent](https://github.com/microsoft/agentsleague-techconnect/issues/46) | 🎨 | [@akshaydmaq](https://github.com/akshaydmaq) | [repo](https://github.com/akshaydmaq/PowerPoint-Reviewer-Agent) | 62.8 | 2026-02-13 | | 37 | [#20](https://github.com/microsoft/agentsleague-techconnect/issues/20) | [Pokémon Memory Game](https://github.com/microsoft/agentsleague-techconnect/issues/20) | 🎨 | [@PriyamvadaSharma](https://github.com/PriyamvadaSharma) | [repo](https://github.com/PriyamvadaSharma/HackFest) | 62.2 | 2026-02-13 | | 38 | [#14](https://github.com/microsoft/agentsleague-techconnect/issues/14) | [Windows Cortex](https://github.com/microsoft/agentsleague-techconnect/issues/14) | 🎨 | [@jrubiosainz, rag2111](https://github.com/jrubiosainz, rag2111) (jrubiosainz, rarocha) | [repo](https://github.com/jrubiosainz/windows-cortex) | 61.1 | 2026-02-13 | | 39 | [#56](https://github.com/microsoft/agentsleague-techconnect/issues/56) | [Compliment my pet](https://github.com/microsoft/agentsleague-techconnect/issues/56) | 💼 | [@tanjaquaijtaal-ux](https://github.com/tanjaquaijtaal-ux) | [repo](https://github.com/tanjaquaijtaal-ux/compliment-my-pet/) | 60.1 | 2026-02-13 | | 40 | [#26](https://github.com/microsoft/agentsleague-techconnect/issues/26) | [Bridgenote](https://github.com/microsoft/agentsleague-techconnect/issues/26) | 🎨 | [@kbeaver16](https://github.com/kbeaver16) | [repo](https://github.com/kbeaver16/bridgenote) | 60.0 | 2026-02-13 | | 41 | [#11](https://github.com/microsoft/agentsleague-techconnect/issues/11) | [Hybrid DNS Copilot](https://github.com/microsoft/agentsleague-techconnect/issues/11) | 🎨 | [@munieswar](https://github.com/munieswar) (kevinboland) | [repo](https://github.com/munieswar/hybrid-dns-architect) | 59.5 | 2026-02-13 | | 42 | [#28](https://github.com/microsoft/agentsleague-techconnect/issues/28) | [creative 404](https://github.com/microsoft/agentsleague-techconnect/issues/28) | 🎨 | [@nagendramishr](https://github.com/nagendramishr) | [repo](https://github.com/nagendramishr/creative404) | 58.9 | 2026-02-13 | | 43 | [#44](https://github.com/microsoft/agentsleague-techconnect/issues/44) | [Corporate Bullshit Translator](https://github.com/microsoft/agentsleague-techconnect/issues/44) | 💼 | [@Aakanksha346](https://github.com/Aakanksha346) | [repo](https://github.com/Aakanksha346/corporate-bullshit-translator) | 56.6 | 2026-02-13 | | 44 | [#50](https://github.com/microsoft/agentsleague-techconnect/issues/50) | [Email Summarizer](https://github.com/microsoft/agentsleague-techconnect/issues/50) | 💼 | [@akshaydmaq](https://github.com/akshaydmaq) | [repo](https://github.com/akshaydmaq/Email-Summarization) | 56.6 | 2026-02-13 | | 45 | [#19](https://github.com/microsoft/agentsleague-techconnect/issues/19) | [My Awesome Architecture Executive Visualizer](https://github.com/microsoft/agentsleague-techconnect/issues/19) | 🎨 | [@JochenMSFT](https://github.com/JochenMSFT) (joschr) | [repo](https://github.com/JochenMSFT/architecture-executive-visualizer) | 53.3 | 2026-02-13 | | 46 | [#30](https://github.com/microsoft/agentsleague-techconnect/issues/30) | [PolicyShield](https://github.com/microsoft/agentsleague-techconnect/issues/30) | 🎨 | [@onsenturk](https://github.com/onsenturk) | [repo](https://github.com/onsenturk/PolicyShield) | 50.6 | 2026-02-13 | | 47 | [#24](https://github.com/microsoft/agentsleague-techconnect/issues/24) | [STEMProjectGenerator](https://github.com/microsoft/agentsleague-techconnect/issues/24) | 🎨 | [@getAlex](https://github.com/getAlex) | [repo](https://github.com/getalex/STEMProjectGenerator) | 50.0 | 2026-02-13 | | 48 | [#36](https://github.com/microsoft/agentsleague-techconnect/issues/36) | [AgentRob](https://github.com/microsoft/agentsleague-techconnect/issues/36) | 🎨 | [@robrita](https://github.com/robrita) | [repo](https://github.com/robrita/standupbot) | 48.4 | 2026-02-13 | | 49 | [#51](https://github.com/microsoft/agentsleague-techconnect/issues/51) | [[Submission]: Biz-Ops Calendar Agent](https://github.com/microsoft/agentsleague-techconnect/issues/51) | 🎨 | — | [repo](https://github.com/aktsmm/biz-ops-calendar-agent) | 27.8 | 2026-02-13 | | 50 | [#57](https://github.com/microsoft/agentsleague-techconnect/issues/57) | [[Submission]: MSRC Script Generator](https://github.com/microsoft/agentsleague-techconnect/issues/57) | 🎨 | — | — | 27.8 | 2026-02-13 | | 51 | [#17](https://github.com/microsoft/agentsleague-techconnect/issues/17) | [Its magic](https://github.com/microsoft/agentsleague-techconnect/issues/17) | 🎨 | — | — | 27.8 | 2026-02-13 | | 52 | [#10](https://github.com/microsoft/agentsleague-techconnect/issues/10) | [My Hackathon Project](https://github.com/microsoft/agentsleague-techconnect/issues/10) | 🎨 | [@monikadubey](https://github.com/monikadubey) | [repo](https://github.com/moni0403/MyGitRepo) | 21.7 | 2026-02-13 | --- ## 📋 Individual Evaluation Summaries ### #27: [TrendSurf](https://github.com/microsoft/agentsleague-techconnect/issues/27) - **Track**: 🧠 Reasoning Agents - **Score**: 91.5/100 - **Submitter**: [@ethanjenkins1](https://github.com/ethanjenkins1) (hedinulo) - **Team**: hedinulo - **Repo**: [https://github.com/ethanjenkins1/trendsurf](https://github.com/ethanjenkins1/trendsurf) - **Description**: **TrendSurf Copilot** is an AI-powered social media content pipeline for regulated industries (like fintech). You give it a trending topic, and it automatically produces publish-ready posts for LinkedIn, X/Twitter, and Microsoft Teams — all checked against your company's brand policy. **How it work… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (24 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: TrendSurf is a reasoning agents submission scoring 91.5/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #38: [Zavatravel](https://github.com/microsoft/agentsleague-techconnect/issues/38) - **Track**: 🧠 Reasoning Agents - **Score**: 91.5/100 - **Submitter**: [@sureshpaulraj](https://github.com/sureshpaulraj) (surep, senthkumar, makizhne) - **Team**: surep, senthkumar, makizhne - **Repo**: [https://github.com/sureshpaulraj/zavatravel](https://github.com/sureshpaulraj/zavatravel) - **Description**: Three AI agents collaborate in a group chat to generate, review, and publish platform-ready social media content for Zava Travel Inc. - **Creator** | Azure OpenAI (GPT-5.x) | Chain-of-Thought (5 steps) | Drafts content with visible reasoning - **Reviewer** | GitHub Copilot SDK | ReAct (Observe → Th… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (22 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Zavatravel is a reasoning agents submission scoring 91.5/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #61: [Social AI Studio](https://github.com/microsoft/agentsleague-techconnect/issues/61) - **Track**: 🧠 Reasoning Agents - **Score**: 91.5/100 - **Submitter**: [@naoki1213mj](https://github.com/naoki1213mj) - **Repo**: [https://github.com/naoki1213mj/social-ai-studio](https://github.com/naoki1213mj/social-ai-studio) - **Description**: **Social AI Studio** is an AI-powered social media content creation pipeline built on Microsoft Foundry. A single **gpt-5.2 reasoning agent** with **7 integrated tools** autonomously executes a **3-phase reasoning pipeline** (Chain-of-Thought → ReAct → Self-Reflection) to generate platform-optimi… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (42 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Social AI Studio is a reasoning agents submission scoring 91.5/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #52: [HeadlineArt: Where News Meets Canvas](https://github.com/microsoft/agentsleague-techconnect/issues/52) - **Track**: 🧠 Reasoning Agents - **Score**: 91.5/100 - **Submitter**: [@FranGenoa, Ajananth, Lakshmy](https://github.com/FranGenoa, Ajananth, Lakshmy) (frfava, ajananth, lchandran, msadhwani, kumarahu) - **Team**: frfava, ajananth, lchandran, msadhwani, kumarahu - **Repo**: [https://github.com/FranGenoa/HeadlineArt](https://github.com/FranGenoa/HeadlineArt) - **Description**: HeadlineArt is a 7-agent AI pipeline built with Microsoft Agent Framework that scans daily news, curates the most visually compelling stories, designs an art concept, writes Instagram-ready captions, and generates the final artwork using gpt-image-1.5, all autonomously with a built-in quality review… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (10 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: HeadlineArt: Where News Meets Canvas is a reasoning agents submission scoring 91.5/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #54: [Saiten — Multi-Agent Scoring System](https://github.com/microsoft/agentsleague-techconnect/issues/54) - **Track**: 🎨 Creative Apps - **Score**: 90.0/100 - **Submitter**: [@aktsmm](https://github.com/aktsmm) (tatsumiy, nmatsumoto, takuyamasuda) - **Team**: tatsumiy, nmatsumoto, takuyamasuda - **Repo**: [https://github.com/aktsmm/FY26_techconnect_saiten](https://github.com/aktsmm/FY26_techconnect_saiten) - **Description**: A multi-agent scoring system that automatically evaluates all Agents League @ TechConnect hackathon submissions and generates ranking reports — just type `@saiten-orchestrator score all` in VS Code. Designed with **Orchestrator-Workers + Prompt Chaining + Evaluator-Optimizer** patterns, 6 Copilot c… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (10 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Saiten — Multi-Agent Scoring System is a creative apps submission scoring 90.0/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #49: [EasyExpenseAI](https://github.com/microsoft/agentsleague-techconnect/issues/49) - **Track**: 🎨 Creative Apps - **Score**: 90.0/100 - **Submitter**: [@chakras](https://github.com/chakras) (arashinkar,jaideeproy) - **Team**: arashinkar,jaideeproy - **Repo**: [https://github.com/chakras/Easy-Expense-AI](https://github.com/chakras/Easy-Expense-AI) - **Description**: An intelligent expense report application powered by a multi-agent AI pipeline built with Microsoft Semantic Kernel, Azure AI Document Intelligence, and Azure OpenAI. EasyExpenseAI automates the tedious process of expense reporting. Users upload receipt images or PDFs — flight tickets, hotel invoic… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (37 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: EasyExpenseAI is a creative apps submission scoring 90.0/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #58: [StoryCircuit](https://github.com/microsoft/agentsleague-techconnect/issues/58) - **Track**: 🧠 Reasoning Agents - **Score**: 88.0/100 - **Submitter**: [@vrengar](https://github.com/vrengar) - **Repo**: [https://github.com/vrengar/storycircuit](https://github.com/vrengar/storycircuit) - **Description**: StoryCircuit is an AI-powered technical narrative architect that transforms complex technical topics into platform-optimized social media content. Users input casual shorthand requests (e.g., "Agentic DevOps in prod") and the agent generates distinct, professional content tailored for LinkedIn, Twit… - **Strengths**: Clear architecture and API documentation with practical setup instructions, Strong grounding intent using web and file search plus platform-specific outputs, Usable interface and export flow for generated content - **Improvements**: Include concrete benchmark examples that prove output quality gains versus a baseline prompt-only approach, Provide implementation evidence for advanced reasoning patterns beyond prompt claims (e.g., trace logs or evaluator outputs) - **Summary**: StoryCircuit is a well-structured content-generation system with clear multi-platform output goals and good documentation hygiene. It shows solid reasoning-oriented design and useful UX patterns for shorthand prompt expansion. The project is strong overall, with slightly less differentiation and depth than the highest-evidence submissions. ### #32: [SovereignFit](https://github.com/microsoft/agentsleague-techconnect/issues/32) - **Track**: 🎨 Creative Apps - **Score**: 86.1/100 - **Submitter**: [@danieletten](https://github.com/danieletten) (daetten) - **Team**: daetten - **Repo**: [https://github.com/danieletten/SovereignFit](https://github.com/danieletten/SovereignFit) - **Description**: SovereignFit is a multi-agent reasoning pipeline that analyzes workload characteristics and recommends the optimal Azure deployment model. It evaluates 5 models (Public Azure, Azure Local Connected, Hybrid, Edge, Azure Local Disconnected) across 5 dynamically weighted dimensions. The pipeline consis… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (8 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: SovereignFit is a creative apps submission scoring 86.1/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #42: [Zava Smart Assistant](https://github.com/microsoft/agentsleague-techconnect/issues/42) - **Track**: 🎨 Creative Apps - **Score**: 85.6/100 - **Submitter**: [@payton-chou-ms](https://github.com/payton-chou-ms) - **Repo**: [https://github.com/payton-chou-ms/poc-agents-league-techconnect-2026](https://github.com/payton-chou-ms/poc-agents-league-techconnect-2026) - **Description**: Zava Smart Assistant is an AI-powered enterprise incident response agent built on the GitHub Copilot SDK. It orchestrates a cross-regional product outage scenario through a single conversational interface, guiding users through 4 phases: Diagnose, Fix, Verify, and Report. The system uses a novel "Sk… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (9 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Zava Smart Assistant is a creative apps submission scoring 85.6/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #35: [techconnect-policy-compliance-checker](https://github.com/microsoft/agentsleague-techconnect/issues/35) - **Track**: 🧠 Reasoning Agents - **Score**: 85.5/100 - **Submitter**: [@taimoor-janjua](https://github.com/taimoor-janjua) - **Repo**: [https://github.com/taimoor-janjua/techconnect-policy-compliance-checker](https://github.com/taimoor-janjua/techconnect-policy-compliance-checker) - **Description**: - The AI Privacy Notice Copilot is a full-stack MVP (React + TypeScript frontend, .NET 8 backend, Azure AI Foundry) that guides non-technical users through creating GDPR-compliant privacy notices via a friendly, one-question-at-a-time chat experience. - It covers all 13 mandatory GDPR Article 13/14… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance - **Improvements**: — - **Summary**: techconnect-policy-compliance-checker is a reasoning agents submission scoring 85.5/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #60: [BrandComm-agent](https://github.com/microsoft/agentsleague-techconnect/issues/60) - **Track**: 🧠 Reasoning Agents - **Score**: 84.0/100 - **Submitter**: [@joyjeet](https://github.com/joyjeet) - **Repo**: [https://github.com/joyjeet/TechConnectHack/tree/main/agent-webapp](https://github.com/joyjeet/TechConnectHack/tree/main/agent-webapp) - **Description**: AI agent in Foundry creates compliant social posts, grounded in CELA rules, with guardrails, monitoring, and Entra ID sign‑in. - **Strengths**: Comprehensive deployment workflow with azd and documented local dev options, Clear authentication architecture using Entra ID and secure app setup, Practical integration path for compliant social content generation - **Improvements**: Add concrete end-to-end reasoning traces (input, tool calls, intermediate decisions, final output), Document measurable quality outcomes from guardrails and compliance checks across sample runs - **Summary**: BrandComm-agent delivers a practical social content assistant with Entra ID sign-in and Azure Foundry integration, plus a clear deployment path via azd. The project is strong operationally, but the available documentation emphasizes platform setup more than deeply demonstrated reasoning behavior. It remains a solid, enterprise-ready implementation with moderate differentiation. ### #13: [SE Agent](https://github.com/microsoft/agentsleague-techconnect/issues/13) - **Track**: 💼 Enterprise Agents - **Score**: 83.4/100 - **Submitter**: [@Daboostr](https://github.com/Daboostr) - **Repo**: [https://github.com/daboostr/SEDemoAgent](https://github.com/daboostr/SEDemoAgent) - **Description**: My Agent will create demo leads in Dynamics 365 for FSI data model. The agent can be customized to generate leads for any industry with a simple tweak to the prompt. - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, All submission checklist items completed, Strong Technical Implementation - **Improvements**: — - **Summary**: SE Agent is a enterprise agents submission scoring 83.4/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #34: [LearnIQ](https://github.com/microsoft/agentsleague-techconnect/issues/34) - **Track**: 🎨 Creative Apps - **Score**: 82.2/100 - **Submitter**: [@kylednoland](https://github.com/kylednoland) - **Repo**: [https://github.com/kylednoland/LearnIQ](https://github.com/kylednoland/LearnIQ) - **Description**: LearnIQ is a personalized learning companion that connects your professional skills and career role to relevant content from the Microsoft Learn catalog. Instead of manually browsing thousands of modules, you define your skills matrix and role, and LearnIQ surfaces the most relevant learning paths —… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (7 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: LearnIQ is a creative apps submission scoring 82.2/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #15: [Skills Hub](https://github.com/microsoft/agentsleague-techconnect/issues/15) - **Track**: 🎨 Creative Apps - **Score**: 81.7/100 - **Submitter**: [@samueltauil, @rleme](https://github.com/samueltauil, @rleme) (samuel.tauil, rodrigoleme) - **Team**: samuel.tauil, rodrigoleme - **Repo**: [https://github.com/samueltauil/skills-hub](https://github.com/samueltauil/skills-hub) - **Description**: Copilot Skills Hub is a curated catalog of GitHub Copilot skills. Skills are instruction files (SKILL.md) that teach Copilot how to handle specific development tasks — from generating commit messages to performing security audits. - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (9 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Skills Hub is a creative apps submission scoring 81.7/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #31: [Threat Incident Swarm Commander](https://github.com/microsoft/agentsleague-techconnect/issues/31) - **Track**: 🎨 Creative Apps - **Score**: 81.7/100 - **Submitter**: [@dondinulos](https://github.com/dondinulos) (ethanjenkins) - **Team**: ethanjenkins - **Repo**: [https://github.com/dondinulos/matrix-smith-swarm](https://github.com/dondinulos/matrix-smith-swarm) - **Description**: Mirrored from the Matrix, an incident starts as one "Agent Smith". Within minutes, it multiplies across services/endpoints. My console shows the replication graph in real-time, explains the spawn reasons, and coordinates specialized agents "Neo" — triage, diagnosis, mitigation, and comms — so respon… - **Strengths**: Distinct and memorable domain framing that maps swarm behavior to incident operations, Modular TypeScript architecture spanning engine, UI, and MCP server, Good observability affordances through state and telemetry MCP tools - **Improvements**: Add explicit safety boundaries for tool-triggered actions and simulation overrides, Include automated tests for strategy transitions and edge-case state handling - **Summary**: Threat Incident Swarm Commander is a creative simulation with a clear incident-response metaphor and modular architecture. It includes well-defined phases, agent behaviors, and MCP control surfaces that demonstrate thoughtful system design. Reliability evidence is moderate rather than weak, warranting an upward adjustment. ### #47: [Semantic Kernel and AutoGen to MAF Modernizer](https://github.com/microsoft/agentsleague-techconnect/issues/47) - **Track**: 🧠 Reasoning Agents - **Score**: 80.5/100 - **Submitter**: [@misantibanez, valegrajales, eraggi-ar, afrancoc2000](https://github.com/misantibanez, valegrajales, eraggi-ar, afrancoc2000) (michelle.santibanez, valentinag, emilioraggi, anafranco) - **Team**: michelle.santibanez, valentinag, emilioraggi, anafranco - **Repo**: [https://github.com/afrancoc2000/tech-connect-2026-sk-modernizer](https://github.com/afrancoc2000/tech-connect-2026-sk-modernizer) - **Description**: An MCP-exposed AI agent that analyzes code written with Semantic Kernel or AutoGen and modernizes it to Microsoft Agent Framework (MAF). Built to integrate with the [MAF Agent Portal](https://github.com/misantibanez/AI-MAF-AGENT-CREATION) as an MCP tool, or used standalone via CLI/MCP in VS Code. - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (10 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Semantic Kernel and AutoGen to MAF Modernizer is a reasoning agents submission scoring 80.5/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #55: [kube-copilot](https://github.com/microsoft/agentsleague-techconnect/issues/55) - **Track**: 🎨 Creative Apps - **Score**: 80.0/100 - **Submitter**: [@openjny](https://github.com/openjny) - **Repo**: [https://github.com/openjny/kube-copilot](https://github.com/openjny/kube-copilot) - **Description**: `kube-copilot` is a terminal UI (TUI) that lets you manage Kubernetes clusters using natural language. Instead of memorizing complex `kubectl` flags and resource names, you simply describe what you want in plain English and let GitHub Copilot translate it to the right command. Built with the GitHub… - **Strengths**: Clear safety guardrails for destructive cluster operations with explicit confirmation flow, Dual documentation grounding (Kubernetes docs + Microsoft Learn MCP) reduces hallucination risk, Well-structured terminal UX with session resilience and timeout protection - **Improvements**: Add task-level evaluation to measure command accuracy across representative kubectl intents, Expand audit logging and rollback guidance for higher-assurance production use - **Summary**: kube-copilot delivers a compelling natural-language Kubernetes TUI with meaningful operational safeguards. The dual MCP grounding strategy and confirmation-based destructive command flow show practical engineering maturity. It stands out as a strong utility-focused creative app with real reliability considerations. ### #12: [Zava Content Studio — Social Media Content Agent](https://github.com/microsoft/agentsleague-techconnect/issues/12) - **Track**: 🧠 Reasoning Agents - **Score**: 79.5/100 - **Submitter**: [@lordlinus, @ganachan, @hieunhums](https://github.com/lordlinus, @ganachan, @hieunhums) (ssattiraju, ganac, hieunhums) - **Team**: ssattiraju, ganac, hieunhums - **Repo**: [https://github.com/lordlinus/hackfest-fy26](https://github.com/lordlinus/hackfest-fy26) - **Description**: AI Content Studio — A multi-agent social media content creation system powered by Microsoft Agent Framework and Azure Durable Functions. Three specialized AI agents (Ideation, Creator, Reviewer) collaborate using Chain-of-Thought, ReAct, and Self-Reflection reasoning patterns to generate platform-op… - **Strengths**: Clear multi-agent role separation (ideation, creation, review) with orchestration intent, Human-in-the-loop approval and revision loop are well aligned to production workflows, Good use-case fit for social content operations - **Improvements**: Publish a complete README with runnable steps, architecture diagram, and sample orchestration traces, Add objective evaluation metrics for review quality and retry-loop effectiveness - **Summary**: Zava Content Studio describes a robust multi-agent pipeline with durable orchestration and human-in-the-loop approval loops. The concept and architecture are strong, especially for serverless workflow design. Missing repository README content in collected evidence reduces certainty at the top score band. ### #16: [FoundryReleaseScout](https://github.com/microsoft/agentsleague-techconnect/issues/16) - **Track**: 🧠 Reasoning Agents - **Score**: 79.0/100 - **Submitter**: [@JW-Sthlm](https://github.com/JW-Sthlm) (jwallquist) - **Team**: jwallquist - **Repo**: [https://github.com/JW-Sthlm/jw-hackfest/](https://github.com/JW-Sthlm/jw-hackfest/) - **Description**: Enterprise AI Release Scouting Agent helps enterprise architects and partners stay current on Microsoft AI platform updates without marketing noise. The agent treats Microsoft Foundry blog posts as signals of change, then verifies each capability against authoritative Microsoft Learn documentation b… - **Strengths**: Well-defined reasoning workflow with explicit detect-verify-evaluate-filter-publish stages, Strong grounding principle using Microsoft Learn as authoritative source, Clear enterprise communication positioning with deterministic output constraints - **Improvements**: Include concrete run artifacts showing excluded unsupported claims and final approved outputs, Add automated regression checks for source quality and citation consistency - **Summary**: FoundryReleaseScout is a focused reasoning agent that prioritizes verifiable enterprise signal over announcement noise. Its verification-first design and hard constraints improve trustworthiness for professional communication outputs. The project shows stronger technical maturity than the baseline reflected. ### #39: [CodeIntel](https://github.com/microsoft/agentsleague-techconnect/issues/39) - **Track**: 🎨 Creative Apps - **Score**: 77.8/100 - **Submitter**: [@Mazher-Ul-Haq](https://github.com/Mazher-Ul-Haq) - **Repo**: [https://github.com/Mazher-Ul-Haq/codeintel](https://github.com/Mazher-Ul-Haq/codeintel) - **Description**: CodeIntel — Team Intelligence Hub CodeIntel transforms raw GitHub repository data and Microsoft 365 signals into actionable team intelligence. It combines an MCP server (8 tools for GitHub Copilot), a REST API, a real-time dashboard, and a CLI — all built in under 100 minutes using GitHub Copilot.… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (5 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: CodeIntel is a creative apps submission scoring 77.8/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #18: [My Proposal Develop](https://github.com/microsoft/agentsleague-techconnect/issues/18) - **Track**: 🧠 Reasoning Agents - **Score**: 77.0/100 - **Submitter**: [@balakreshnan](https://github.com/balakreshnan) - **Repo**: [https://github.com/balakreshnan/TechconnectFY26Hackfest](https://github.com/balakreshnan/TechconnectFY26Hackfest) - **Description**: My idea is to build agent for RFP proposal with foundry IQ and then make it a CI/CD to deploy to various environments - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance - **Improvements**: — - **Summary**: My Proposal Develop is a reasoning agents submission scoring 77.0/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #43: [Nimbus Content Agent](https://github.com/microsoft/agentsleague-techconnect/issues/43) - **Track**: 🧠 Reasoning Agents - **Score**: 76.5/100 - **Submitter**: [@namratasherkhane](https://github.com/namratasherkhane) - **Repo**: [https://github.com/namratasherkhane/namratasherkhane](https://github.com/namratasherkhane/namratasherkhane) - **Description**: AI agent that helps a loan-focused BFSI comms team craft platform-ready social content. It uses Microsoft Foundry for reasoning, grounds on brand and industry context, and can be extended with MCP tools for actions like saving drafts or fetching news. This project is a BFSI social content agent buil… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (7 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Nimbus Content Agent is a reasoning agents submission scoring 76.5/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #23: [repo-analyzer](https://github.com/microsoft/agentsleague-techconnect/issues/23) - **Track**: 🎨 Creative Apps - **Score**: 76.1/100 - **Submitter**: [@olivomarco](https://github.com/olivomarco) - **Repo**: [https://github.com/olivomarco/repo-analyzer](https://github.com/olivomarco/repo-analyzer) - **Description**: **Repo Inspector** analyzes any GitHub repository and provides actionable insights across **people**, **code quality**, **security**, and **team dynamics**. It combines deterministic metrics with LLM-powered intelligence to surface risks, identify knowledge silos, and help teams make better decision… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance - **Improvements**: — - **Summary**: repo-analyzer is a creative apps submission scoring 76.1/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #53: [Marketing Asset Creation Agent](https://github.com/microsoft/agentsleague-techconnect/issues/53) - **Track**: 🎨 Creative Apps - **Score**: 75.6/100 - **Submitter**: [@jetzlstorfer @stephan-nik](https://github.com/jetzlstorfer @stephan-nik) (jetzlstorfer, sniklas) - **Team**: jetzlstorfer, sniklas - **Repo**: [https://github.com/jetzlstorfer/marketing-asset-creation-agent](https://github.com/jetzlstorfer/marketing-asset-creation-agent) - **Description**: Conference and meetup organizers struggle with incomplete, scattered, and outdated speaker profiles. Speakers submit partial bios late, and their information is spread across LinkedIn, GitHub, Twitter, and personal sites — delaying promotional efforts. This project solves that with a multi-agent AI… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (5 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Marketing Asset Creation Agent is a creative apps submission scoring 75.6/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #29: [Upgrade Sage](https://github.com/microsoft/agentsleague-techconnect/issues/29) - **Track**: 🧠 Reasoning Agents - **Score**: 74.5/100 - **Submitter**: [@betanu701](https://github.com/betanu701) - **Repo**: [https://github.com/Betanu701/UpgradeSage](https://github.com/Betanu701/UpgradeSage) - **Description**: Many companies rely on legacy projects that lag several versions behind modern releases. When organizations finally attempt to upgrade, they often face a long, complicated migration path filled with unknown breaking changes and hidden risks. UpgradeSage solves this problem by analyzing differences b… - **Strengths**: Clear two-tier architecture with dedicated backend analysis and frontend reporting, Concrete developer workflow around version-to-version diff analysis, Real-time interaction model improves usability for upgrade planning - **Improvements**: Add validation benchmarks against known breaking-change datasets to calibrate risk scoring quality, Provide deeper examples of multi-hop upgrade path planning beyond single from/to comparisons - **Summary**: Upgrade Sage provides a useful migration-risk analysis workflow that combines code-diff extraction with AI-generated mitigation guidance. The architecture is practical and includes a functional UI plus backend service separation. It is more complete technically than baseline indicated, while still early-stage in evaluation rigor. ### #37: [Making money by poop](https://github.com/microsoft/agentsleague-techconnect/issues/37) - **Track**: 🎨 Creative Apps - **Score**: 73.9/100 - **Submitter**: [@bensonwugit](https://github.com/bensonwugit) (benswu, YENJUSU, thhuan, huihuan, yumwang) - **Team**: benswu, YENJUSU, thhuan, huihuan, yumwang - **Repo**: [https://github.com/bensonwugit/make_money_by_shit](https://github.com/bensonwugit/make_money_by_shit) - **Description**: Start making money while pooping at work! 🚽💰 Ever wondered how much you've earned just by sitting on the toilet? This hilarious yet powerful app lets you track every poop, calculate your exact earnings, and optimize your bathroom breaks for maximum profit. Because let's face it—your time is money, a… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance - **Improvements**: — - **Summary**: Making money by poop is a creative apps submission scoring 73.9/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #48: [CommsGuard](https://github.com/microsoft/agentsleague-techconnect/issues/48) - **Track**: 🧠 Reasoning Agents - **Score**: 73.5/100 - **Submitter**: [@Akshay8870](https://github.com/Akshay8870) (sankhac , jaideeproy) - **Team**: sankhac , jaideeproy - **Repo**: [https://github.com/Akshay8870/bfsi-commsguard.git](https://github.com/Akshay8870/bfsi-commsguard.git) - **Description**: CommsGuard AI is a multi-agent reasoning system built on Microsoft Foundry to help BFSI organizations create compliant, high-quality social media content. The system orchestrates four specialized agents: a Planner that decomposes campaign objectives into structured tasks, a Writer that generates pla… - **Strengths**: Strong domain fit for regulated BFSI communications, Clear compliance-first framing with policy-gated publishing intent, Includes demo artifacts indicating practical workflow execution - **Improvements**: Add a complete README with architecture, setup, and concrete tool flow to support score confidence, Provide reproducible examples showing blocked unsafe outputs and approved compliant rewrites - **Summary**: CommsGuard presents a compelling BFSI compliance use case with planner/writer/reviewer style orchestration and safety intent. However, absent README content in the collected dataset limits verifiable evidence for top-tier scoring. The submission remains promising but currently under-documented for higher confidence evaluation. ### #33: [BeatMyPrompt](https://github.com/microsoft/agentsleague-techconnect/issues/33) - **Track**: 🎨 Creative Apps - **Score**: 72.8/100 - **Submitter**: [@fimdim](https://github.com/fimdim) - **Repo**: [https://github.com/fimdim/BeatMyPrompt](https://github.com/fimdim/BeatMyPrompt) - **Description**: **BeatMyPrompt** is a live AI battle experience where two AI‑generated verses go head‑to‑head, and the **audience decides the winner using their applause**. No apps, no sign‑ups, no friction. Just: - AI for **creative generation** - A crowd for **judgment** - And a big, visual **Clap‑O‑Meter** th… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (15 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: BeatMyPrompt is a creative apps submission scoring 72.8/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #22: [DailySync](https://github.com/microsoft/agentsleague-techconnect/issues/22) - **Track**: 🎨 Creative Apps - **Score**: 71.1/100 - **Submitter**: [@robertoborges](https://github.com/robertoborges) (roborges) - **Team**: roborges - **Repo**: [https://github.com/RobertoBorges/techconnect-Hackfest](https://github.com/RobertoBorges/techconnect-Hackfest) - **Description**: AI-powered CLI tool that auto-generates daily standup reports by pulling real developer activity from GitHub (commits, PRs, issues) and Microsoft 365 (calendar, Teams, action items) via MCP. Saves ~10 min/day writing standups. - **Strengths**: Comprehensive README documentation, Public repository available, Rich technology stack (9 technologies), All submission checklist items completed, Strong Accuracy & Relevance - **Improvements**: No demo materials provided, Weak Reasoning & Multi-step Thinking - **Summary**: DailySync is a creative apps submission scoring 71.1/100. Key strengths: Comprehensive README documentation, Public repository available. Areas for improvement: No demo materials provided, Weak Reasoning & Multi-step Thinking. ### #41: [Architect Agent](https://github.com/microsoft/agentsleague-techconnect/issues/41) - **Track**: 🎨 Creative Apps - **Score**: 69.5/100 - **Submitter**: [@bzbrkic](https://github.com/bzbrkic) - **Repo**: [https://github.com/bzbrkic/Agent-Architect](https://github.com/bzbrkic/Agent-Architect) - **Description**: Agent Architect helps teams determine the best approach for building AI agents by analyzing customer needs, technical constraints, and requirements. It provides build vs. buy recommendations, architecture outlines, and starter prompts for implementation. - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (16 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Architect Agent is a creative apps submission scoring 69.5/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #40: [AI Project Recap Storyboard](https://github.com/microsoft/agentsleague-techconnect/issues/40) - **Track**: 🎨 Creative Apps - **Score**: 66.1/100 - **Submitter**: [@huier23, dinowang, chhtw, Phonicdream](https://github.com/huier23, dinowang, chhtw, Phonicdream) (huihuan, yumwang, thhuan, yenjusu) - **Team**: huihuan, yumwang, thhuan, yenjusu - **Repo**: [https://github.com/where-is-my-org/Creative-Apps](https://github.com/where-is-my-org/Creative-Apps) - **Description**: Recap Storyboard turns GitHub activity and local notes into a narrative recap with chapters, highlights, risks, and a timeline. It pulls PRs and commits for a chosen date range, blends in local notes that capture decisions, risks, and next steps, and generates a recap payload for a storyboard-style… - **Strengths**: Demo video or screenshots provided, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance, Strong Reasoning & Multi-step Thinking - **Improvements**: — - **Summary**: AI Project Recap Storyboard is a creative apps submission scoring 66.1/100. Key strengths: Demo video or screenshots provided, Public repository available. ### #59: [Dictation 4 Kids](https://github.com/microsoft/agentsleague-techconnect/issues/59) - **Track**: 🎨 Creative Apps - **Score**: 66.1/100 - **Submitter**: [@kevintkliu](https://github.com/kevintkliu) (kevinliu2, erikatang) - **Team**: kevinliu2, erikatang - **Repo**: [https://github.com/kevintkliu/dictation4kids](https://github.com/kevintkliu/dictation4kids) - **Description**: 🌟 Introduction to Dictation 4 Kids As a parent, helping our children study can be both rewarding and… a little challenging. When I help my son prepare for dictation, he often asks the same questions many kids do: “What does this word mean?” “Can you read it for me again?” “How do I pronounce it?” “… - **Strengths**: Demo video or screenshots provided, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance, Strong UX & Presentation - **Improvements**: No README found or repository inaccessible - **Summary**: Dictation 4 Kids is a creative apps submission scoring 66.1/100. Key strengths: Demo video or screenshots provided, Public repository available. Areas for improvement: No README found or repository inaccessible. ### #25: [Team Risk Analyzer](https://github.com/microsoft/agentsleague-techconnect/issues/25) - **Track**: 🎨 Creative Apps - **Score**: 65.6/100 - **Submitter**: [@rebmid](https://github.com/rebmid) - **Repo**: [https://github.com/rebmid/team-risk-analyzer](https://github.com/rebmid/team-risk-analyzer) - **Description**: Team Risk Analyzer is an MCP‑enabled CLI tool that helps teams keep a pulse on their day‑to‑day delivery health inside VS Code. It reviews pull requests, issues, and meeting notes—using the same style of metadata you’d find in GitHub—to spot potential risks early. That includes things like aging PRs… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (5 technologies), All submission checklist items completed - **Improvements**: — - **Summary**: Team Risk Analyzer is a creative apps submission scoring 65.6/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. ### #21: [Deal to Glory](https://github.com/microsoft/agentsleague-techconnect/issues/21) - **Track**: 🎨 Creative Apps - **Score**: 63.9/100 - **Submitter**: [@billwhalenmsft](https://github.com/billwhalenmsft) - **Repo**: [https://github.com/billwhalenmsft/agents-league-deal-to-glory**](https://github.com/billwhalenmsft/agents-league-deal-to-glory**) - **Description**: Deal to Glory is a voice-first, multi-agent workflow that turns a seller's post-meeting sentence into a closed-loop CRM update, commission calculation, and manager recognition moment — all in under 5 seconds. A seller says: "Just finished a call with Acme. Deal is now committed at 420K, closing mid… - **Strengths**: Compelling end-to-end sales scenario from update capture to recognition and notifications, Clear multi-agent stage decomposition aligned to real business workflow steps, Good UX ambition with live visualization and what-if forecasting - **Improvements**: Fix repository metadata/access and provide complete technical documentation to support reproducibility, Add robust failure-mode handling details for speech parsing, CRM updates, and notification delivery - **Summary**: Deal to Glory proposes an engaging voice-first sales workflow with clear multi-agent stages and manager notification outcomes. The business narrative is strong, but verifiable repository documentation is currently limited in collected data. Score is adjusted to reflect concept strength with reduced implementation confidence. ### #45: [Prompt Escape - The AI Escape Room Craft the right prompts. Solve the puzzles. Escape.](https://github.com/microsoft/agentsleague-techconnect/issues/45) - **Track**: 🎨 Creative Apps - **Score**: 63.9/100 - **Submitter**: [@billwhalenmsft](https://github.com/billwhalenmsft) - **Repo**: [https://github.com/billwhalenmsft/PromptEscape](https://github.com/billwhalenmsft/PromptEscape) - **Description**: Prompt Escape is a visual, prompt-engineering escape room game where players write AI prompts to solve puzzles and escape themed rooms. It features 3 themed adventures — Indiana Jones, Harry Potter, and ClippyLand — with CSS art, animations, and a retro-gaming aesthetic. Built with Node.js + Express… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, Rich technology stack (9 technologies), All submission checklist items completed - **Improvements**: Weak Reasoning & Multi-step Thinking, Weak Reliability & Safety - **Summary**: Prompt Escape - The AI Escape Room Craft the right prompts. Solve the puzzles. Escape. is a creative apps submission scoring 63.9/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. Areas for improvement: Weak Reasoning & Multi-step Thinking, Weak Reliability & Safety. ### #46: [PowerPoint Reviewer Agent](https://github.com/microsoft/agentsleague-techconnect/issues/46) - **Track**: 🎨 Creative Apps - **Score**: 62.8/100 - **Submitter**: [@akshaydmaq](https://github.com/akshaydmaq) - **Repo**: [https://github.com/akshaydmaq/PowerPoint-Reviewer-Agent](https://github.com/akshaydmaq/PowerPoint-Reviewer-Agent) - **Description**: An AI-powered agent that automatically reviews and corrects PowerPoint presentations using OpenAI GPT-4. Features Spelling & Grammar Correction - AI-powered detection and fixing of errors Alignment Standardization - Automatically aligns titles and elements consistently Context-Aware Decisions - Use… - **Strengths**: Demo video or screenshots provided, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance, Strong Creativity & Originality - **Improvements**: Weak Reasoning & Multi-step Thinking - **Summary**: PowerPoint Reviewer Agent is a creative apps submission scoring 62.8/100. Key strengths: Demo video or screenshots provided, Public repository available. Areas for improvement: Weak Reasoning & Multi-step Thinking. ### #20: [Pokémon Memory Game](https://github.com/microsoft/agentsleague-techconnect/issues/20) - **Track**: 🎨 Creative Apps - **Score**: 62.2/100 - **Submitter**: [@PriyamvadaSharma](https://github.com/PriyamvadaSharma) - **Repo**: [https://github.com/PriyamvadaSharma/HackFest](https://github.com/PriyamvadaSharma/HackFest) - **Description**: A React-based memory card game built for the TechConnect Hack Fest Creative Apps challenge. Players flip cards to find matching Pokemon pairs from a 4x4 grid featuring 8 iconic Pokemon characters (Pikachu, Charizard, Dragonite, and more). Key Features: Single and multiplayer modes (1-2 players) S… - **Strengths**: Demo video or screenshots provided, Public repository available, Rich technology stack (9 technologies), All submission checklist items completed, Strong Accuracy & Relevance - **Improvements**: No README found or repository inaccessible, Weak Reliability & Safety - **Summary**: Pokémon Memory Game is a creative apps submission scoring 62.2/100. Key strengths: Demo video or screenshots provided, Public repository available. Areas for improvement: No README found or repository inaccessible, Weak Reliability & Safety. ### #14: [Windows Cortex](https://github.com/microsoft/agentsleague-techconnect/issues/14) - **Track**: 🎨 Creative Apps - **Score**: 61.1/100 - **Submitter**: [@jrubiosainz, rag2111](https://github.com/jrubiosainz, rag2111) (jrubiosainz, rarocha) - **Team**: jrubiosainz, rarocha - **Repo**: [https://github.com/jrubiosainz/windows-cortex](https://github.com/jrubiosainz/windows-cortex) - **Description**: My app is a Claude Cowork Killer... using 100% Microsoft tools, accessible for anyone through text or voice, and running on Windows! Windows Cortex is a native desktop application that gives you AI superpowers over your Windows PC. Just type (or speak) what you want to do in natural language, and t… - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance - **Improvements**: Weak Reasoning & Multi-step Thinking, Weak Reliability & Safety - **Summary**: Windows Cortex is a creative apps submission scoring 61.1/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. Areas for improvement: Weak Reasoning & Multi-step Thinking, Weak Reliability & Safety. ### #56: [Compliment my pet](https://github.com/microsoft/agentsleague-techconnect/issues/56) - **Track**: 💼 Enterprise Agents - **Score**: 60.1/100 - **Submitter**: [@tanjaquaijtaal-ux](https://github.com/tanjaquaijtaal-ux) - **Repo**: [https://github.com/tanjaquaijtaal-ux/compliment-my-pet/](https://github.com/tanjaquaijtaal-ux/compliment-my-pet/) - **Description**: >**AI doesn’t always have to be productive. >Sometimes, it just needs to make people smile.** **Compliment My Pet** is a feel‑good AI agent optimized for joy, not complexity. Its purpose is simple: boost people’s mood by letting them talk about what they love most — their pets. Users can upload… - **Strengths**: Clear UX intent and coherent prompt design for positive conversational outcomes, Good onboarding guidance for Copilot Studio import and activation, Demonstrates awareness of responsible AI behavior and real-world limitations - **Improvements**: Quantify measurable enterprise outcomes (e.g., engagement, retention, wellbeing metrics) to justify business impact, Add broader enterprise scenarios beyond pet-focused conversation to improve scalability - **Summary**: Compliment my pet is a polished, human-centered conversational experience with clear emotional design intent. It demonstrates practical Copilot Studio packaging and deployment steps. For enterprise scoring, business impact is more limited than workflow-focused agents, so value is moderated accordingly. ### #26: [Bridgenote](https://github.com/microsoft/agentsleague-techconnect/issues/26) - **Track**: 🎨 Creative Apps - **Score**: 60.0/100 - **Submitter**: [@kbeaver16](https://github.com/kbeaver16) - **Repo**: [https://github.com/kbeaver16/bridgenote](https://github.com/kbeaver16/bridgenote) - **Description**: BridgeNote — Project Description BridgeNote is a creative note‑transformation assistant designed for people who work across disconnected systems and constantly email themselves notes just to keep information in sync. Instead of juggling inboxes or manually reformatting content for OneNote or other t… - **Strengths**: Practical workflow pain point and clear value proposition for cross-system note handling, Multi-format output concept (summary, actions, rewrite, simplified explanation) is user-friendly, Reasonable extensibility direction via MCP integrations - **Improvements**: Provide a complete README with architecture, dependencies, and clean cross-platform setup instructions, Replace machine-specific local paths with reproducible commands and validated quickstart - **Summary**: BridgeNote proposes a useful note-transformation assistant concept with multiple output modes and optional context enrichment. However, collected evidence lacks a retrievable README and clear reproducible setup flow, limiting verification of implementation depth. The idea is promising but currently under-documented. ### #11: [Hybrid DNS Copilot](https://github.com/microsoft/agentsleague-techconnect/issues/11) - **Track**: 🎨 Creative Apps - **Score**: 59.5/100 - **Submitter**: [@munieswar](https://github.com/munieswar) (kevinboland) - **Team**: kevinboland - **Repo**: [https://github.com/munieswar/hybrid-dns-architect](https://github.com/munieswar/hybrid-dns-architect) - **Description**: Hybrid DNS Architect helps network engineers and cloud architects visualize, understand, and deploy Azure DNS Private Resolver configurations for hybrid DNS environments. It tackles the complexity of setting up bidirectional DNS resolution between Azure and on-premises networks. Interactive Archite… - **Strengths**: Comprehensive README documentation, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance, Strong UX & Presentation - **Improvements**: No demo materials provided, Weak Reasoning & Multi-step Thinking, Weak Reliability & Safety - **Summary**: Hybrid DNS Copilot is a creative apps submission scoring 59.5/100. Key strengths: Comprehensive README documentation, Public repository available. Areas for improvement: No demo materials provided, Weak Reasoning & Multi-step Thinking. ### #28: [creative 404](https://github.com/microsoft/agentsleague-techconnect/issues/28) - **Track**: 🎨 Creative Apps - **Score**: 58.9/100 - **Submitter**: [@nagendramishr](https://github.com/nagendramishr) - **Repo**: [https://github.com/nagendramishr/creative404](https://github.com/nagendramishr/creative404) - **Description**: Generates a 404 image for use with websites. As input, you point the generator to a working MCP server. The code will inspect the server and based on the capabilities, generate a graphic that is apropriate. - **Strengths**: Demo video or screenshots provided, Comprehensive README documentation, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance - **Improvements**: Weak Reasoning & Multi-step Thinking - **Summary**: creative 404 is a creative apps submission scoring 58.9/100. Key strengths: Demo video or screenshots provided, Comprehensive README documentation. Areas for improvement: Weak Reasoning & Multi-step Thinking. ### #44: [Corporate Bullshit Translator](https://github.com/microsoft/agentsleague-techconnect/issues/44) - **Track**: 💼 Enterprise Agents - **Score**: 56.6/100 - **Submitter**: [@Aakanksha346](https://github.com/Aakanksha346) - **Repo**: [https://github.com/Aakanksha346/corporate-bullshit-translator](https://github.com/Aakanksha346/corporate-bullshit-translator) - **Description**: Corporate Bullshit Translator is an intelligent M365 Copilot agent that analyzes meeting transcripts and corporate communications to identify vague business jargon and translate it into clear, honest language. The agent calculates a "Bullshit Score" (1-10) based on buzzword density and provides… - **Strengths**: Strong and relatable business pain point around unclear corporate communication, Creative framing that can improve meeting clarity and decision speed, Potential for practical adoption in communication-heavy teams - **Improvements**: Add full technical documentation (agent architecture, data flow, authentication, and deployment), Provide a working demo trace with input transcript, computed score, and generated recommendations - **Summary**: Corporate Bullshit Translator has a memorable enterprise communication use case and a clear problem narrative. However, repository documentation is currently too sparse to validate advanced technical architecture claims. The concept is useful, but implementation evidence needs to catch up for higher confidence scoring. ### #50: [Email Summarizer](https://github.com/microsoft/agentsleague-techconnect/issues/50) - **Track**: 💼 Enterprise Agents - **Score**: 56.6/100 - **Submitter**: [@akshaydmaq](https://github.com/akshaydmaq) - **Repo**: [https://github.com/akshaydmaq/Email-Summarization](https://github.com/akshaydmaq/Email-Summarization) - **Description**: An agent that reads the user's email inbox, summarizes emails, and identifies action items, prioritizing emails from customers list , low sentiment emails, and escalation emails for bugs and issues reported. This will help us to focus on what is most important/urgent mail to reply. - **Strengths**: Clear enterprise scenario with immediate applicability to support and operations teams, Straightforward user flow for identifying high-priority communications, Includes demo material despite privacy constraints - **Improvements**: Expand technical documentation with architecture, data handling, and security controls, Add evaluation results showing precision/recall of urgent-email detection against labeled samples - **Summary**: Email Summarizer targets a practical enterprise need: prioritizing inbox items by urgency and actionability. The implementation appears lightweight but directionally useful for operational efficiency. Scope is narrow, yet business relevance is clearer than the baseline reflected. ### #19: [My Awesome Architecture Executive Visualizer](https://github.com/microsoft/agentsleague-techconnect/issues/19) - **Track**: 🎨 Creative Apps - **Score**: 53.3/100 - **Submitter**: [@JochenMSFT](https://github.com/JochenMSFT) (joschr) - **Team**: joschr - **Repo**: [https://github.com/JochenMSFT/architecture-executive-visualizer](https://github.com/JochenMSFT/architecture-executive-visualizer) - **Description**: My agent can automatically generate PowerPoint slides from technical artifacts. - **Strengths**: Demo video or screenshots provided, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance, Strong UX & Presentation - **Improvements**: No README found or repository inaccessible, Weak Reasoning & Multi-step Thinking, Weak Reliability & Safety - **Summary**: My Awesome Architecture Executive Visualizer is a creative apps submission scoring 53.3/100. Key strengths: Demo video or screenshots provided, Public repository available. Areas for improvement: No README found or repository inaccessible, Weak Reasoning & Multi-step Thinking. ### #30: [PolicyShield](https://github.com/microsoft/agentsleague-techconnect/issues/30) - **Track**: 🎨 Creative Apps - **Score**: 50.6/100 - **Submitter**: [@onsenturk](https://github.com/onsenturk) - **Repo**: [https://github.com/onsenturk/PolicyShield](https://github.com/onsenturk/PolicyShield) - **Description**: PolicyShield is a Next.js web app that enables enterprise policy-as-code. Regulations and controls evolve, so instead of relying on static built-in policies, customers can generate and own their compliance bundles. Users select a top regulation (or type one), Azure OpenAI extracts control requiremen… - **Strengths**: Strong problem framing around policy-as-code and compliance ownership, Useful output direction with multi-format IaC bundle generation, Clear intention for transparent AI-assisted mapping and coverage reporting - **Improvements**: Add full README with architecture, setup, sample inputs/outputs, and deployment validation, Document error handling, input validation, and security controls for generated compliance artifacts - **Summary**: PolicyShield targets a relevant enterprise compliance challenge with policy-bundle generation and coverage analysis. The concept is practical, but available repository documentation in collected data is insufficient to validate higher UX and reliability scoring. Current assessment reflects potential with limited verifiable evidence. ### #24: [STEMProjectGenerator](https://github.com/microsoft/agentsleague-techconnect/issues/24) - **Track**: 🎨 Creative Apps - **Score**: 50.0/100 - **Submitter**: [@getAlex](https://github.com/getAlex) - **Repo**: [https://github.com/getalex/STEMProjectGenerator](https://github.com/getalex/STEMProjectGenerator) - **Description**: A web solution to generate new STEM project ideas for kids. - **Strengths**: Demo video or screenshots provided, Public repository available, All submission checklist items completed, Strong Accuracy & Relevance, Strong UX & Presentation - **Improvements**: No README found or repository inaccessible, Weak Reasoning & Multi-step Thinking, Weak Creativity & Originality - **Summary**: STEMProjectGenerator is a creative apps submission scoring 50.0/100. Key strengths: Demo video or screenshots provided, Public repository available. Areas for improvement: No README found or repository inaccessible, Weak Reasoning & Multi-step Thinking. ### #36: [AgentRob](https://github.com/microsoft/agentsleague-techconnect/issues/36) - **Track**: 🎨 Creative Apps - **Score**: 48.4/100 - **Submitter**: [@robrita](https://github.com/robrita) - **Repo**: [https://github.com/robrita/standupbot](https://github.com/robrita/standupbot) - **Description**: **Real-time voice-enabled AI Scrum Master** that facilitates daily standup meetings, collects structured updates from each team member, and delivers an actionable digest — powered by Azure Voice Live API and Azure AI Foundry. - **Strengths**: Demo video or screenshots provided, Public repository available, All submission checklist items completed, Strong UX & Presentation - **Improvements**: Weak Reasoning & Multi-step Thinking, Weak Reliability & Safety - **Summary**: AgentRob is a creative apps submission scoring 48.4/100. Key strengths: Demo video or screenshots provided, Public repository available. Areas for improvement: Weak Reasoning & Multi-step Thinking, Weak Reliability & Safety. ### #51: [[Submission]: Biz-Ops Calendar Agent](https://github.com/microsoft/agentsleague-techconnect/issues/51) - **Track**: 🎨 Creative Apps - **Score**: 27.8/100 - **Submitter**: [@aktsmm](https://github.com/aktsmm) - **Repo**: [https://github.com/aktsmm/biz-ops-calendar-agent](https://github.com/aktsmm/biz-ops-calendar-agent) - **Description**: Biz-Ops Calendar Agent is a Copilot Studio agent deployed to M365 Copilot Chat (Teams) that provides smart scheduling and email management through Connected Agents (Calendar Sub-Agent + Email Sub-Agent) and deep Instruction Engineering with structured output formats (★/🔄 markers + ◎○△ point tables).… - **Strengths**: — - **Improvements**: No demo materials provided, No README found or repository inaccessible, No repository URL provided, Weak Accuracy & Relevance, Weak Reasoning & Multi-step Thinking - **Summary**: [Submission]: Biz-Ops Calendar Agent is a creative apps submission scoring 27.8/100. Areas for improvement: No demo materials provided, No README found or repository inaccessible. ### #57: [[Submission]: MSRC Script Generator](https://github.com/microsoft/agentsleague-techconnect/issues/57) - **Track**: 🎨 Creative Apps - **Score**: 27.8/100 - **Strengths**: — - **Improvements**: No demo materials provided, No README found or repository inaccessible, No repository URL provided, Weak Accuracy & Relevance, Weak Reasoning & Multi-step Thinking - **Summary**: [Submission]: MSRC Script Generator is a creative apps submission scoring 27.8/100. Areas for improvement: No demo materials provided, No README found or repository inaccessible. ### #17: [Its magic](https://github.com/microsoft/agentsleague-techconnect/issues/17) - **Track**: 🎨 Creative Apps - **Score**: 27.8/100 - **Strengths**: — - **Improvements**: No demo materials provided, No README found or repository inaccessible, No repository URL provided, Weak Accuracy & Relevance, Weak Reasoning & Multi-step Thinking - **Summary**: Its magic is a creative apps submission scoring 27.8/100. Areas for improvement: No demo materials provided, No README found or repository inaccessible. ### #10: [My Hackathon Project](https://github.com/microsoft/agentsleague-techconnect/issues/10) - **Track**: 🎨 Creative Apps - **Score**: 21.7/100 - **Submitter**: [@monikadubey](https://github.com/monikadubey) - **Repo**: [https://github.com/moni0403/MyGitRepo](https://github.com/moni0403/MyGitRepo) - **Description**: My Agent is just to get first prize for submission which actually does nothing - **Strengths**: Submission was completed and includes a public repository link, Demo artifact is attached in the issue thread - **Improvements**: Implement a working MVP aligned to the chosen track requirements, Add a real README with setup steps, architecture, and measurable functionality - **Summary**: My Hackathon Project is effectively a placeholder submission without demonstrated functionality. The issue text explicitly notes that the project does not do anything, and repository documentation does not provide implementation evidence. Score is reduced to align with rubric minimums for non-functional entries.