Pinned Loading

1. Hanamaruki-Approach-Method-TPS-AI-Engineering-Observation-Framework Hanamaruki-Approach-Method-TPS-AI-Engineering-Observation-Framework Public

   1

2. AI-Kyoji-Sekkei-Unified-Kernel AI-Kyoji-Sekkei-Unified-Kernel Public

   本プロジェクトは、人間がAIを教え、AIが人間に思考を教え返す―― その「共進化の構造」を実装したオープンソース知能設計体系です。note.comの記事 https://note.com/cheeky_daisy8872/n/nc76c7cdc6398?app_launch=false

   1

3. AI_Agent_Risk_Shield-A-User-Safety-Framework-Preventing-Runaway-Costs AI_Agent_Risk_Shield-A-User-Safety-Framework-Preventing-Runaway-Costs Public

   A bilingual safety framework protecting users from runaway AI agents, token explosions, hidden background tasks, and catastrophic billing risks across Gemini, OpenAI, Claude, Grok, and Qwen.**

   1

4. AI_Agent_Safety_Standard-TPS_Based-Minimal_Cost-Solution-for-API-Agent-Economic-Risk AI_Agent_Safety_Standard-TPS_Based-Minimal_Cost-Solution-for-API-Agent-Economic-Risk Public

   A TPS-based minimal-cost safety standard that prevents API×Agent economic failures by introducing the U-Passkey mechanism. Simple, universal, and benefits governments, AI companies, and users simul…

   1

5. GPT5.0-Impact-Report-by-Hanamaruki GPT5.0-Impact-Report-by-Hanamaruki Public

   A user report on the unintended model switching and cognitive regression in ChatGPT 5.0 「Endorsed by DeepMind COO Lila Ibrahim: SOVOS for AI Safety—Cognitive Contamination Proof」

   1

6. TPS-Driven-API-Resilience-Architecture TPS-Driven-API-Resilience-Architecture Public

   TPS-Driven API Resilience Architecture — A Dual-Line Design for Stable Legacy Systems and Scalable Next-Generation APIs

   1