AI & Data•Aug 16, 2026•11 min readZH·ENClaude 开始给文字打水印：原理藏在模型犹豫的地方，而「只有欧盟在管」是个误会Anthropic 说水印不加字符、不加 token、不加钱、读者看不出来——那它到底改了什么？答案是模型挑词的那一刻。顺便把「只有欧盟强制水印」这个印象核了一遍：中国早了 11 个月，加州跟欧盟同一天。ClaudeAnthropic水印SynthIDEU AI Act监管欧洲C2PA熵
AI & Data•Aug 05, 2026•10 min readEN·ZHThis Summer I Made Kimi K3 My Primary ModelKimi K3, Qwen 3.8 Max, and the GA release of DeepSeek V4 Flash all shipped within a month. GLM-5.3 and Kimi K3.1 are still rumors. I also switched my daily-driver model from Claude Code Opus to Kimi K3. This post fact-checks the wave and lays out what actually changed for me.Kimi K3QwenQwen 3.8 MaxDeepSeekDeepSeek V4 FlashGLM-5.3ClaudeClaude OpusClaude Fable 5Moonshot AIAlibabaZhipuOpen Source Models
AI & Data•Aug 03, 2026•10 min readZH·EN我去验证「Codex 不会自省」，最后发现是我自己的程序先挂了电话一篇博文说 Codex 的主 session 没有纠错的自省能力，Claude 才有，还建议 context 别超过 300k。我刚把 Kimi K3 接进自己的 Web Claude Code Pilot，正想知道它能不能当 orchestrator，就去验了一遍。k3 和 Sol 都有自省能力；那条 300k 的建议也是空的，因为 Codex 在 258.4k 就自己压缩了。而「Codex 主 session 没产出」的症状，是我自己的客户端提前关流造成的——那个 bug 只长在 Codex 那条通道上。AI AgentsCodexGPT-5.6Kimi K3Claudesubagentorchestrator上下文窗口Web Claude Code Pilot
AI & Data•Jun 28, 2026•10 min readENA 0-Star Skill Ranked Third on the PPT Leaderboard. So I Stopped Reading It and Built the Decks Myself.A leaderboard scored six Claude 'PPT skills' by feature blurb and ranked a 0-star project above one with 23,000 stars. So I gave the top tools the same real client brief and built the decks. The ranking didn't survive contact with the output.AI AgentsClaudeClaude CodeAgent SkillsPPTPowerPointpptxbaoyu-designAWSDeveloper Tools
AI & Data•May 31, 2026•9 min readENI Let Opus 4.8 Build a Workflow. It Spawned 77 Agents and Burned Five Hours of Quota in Ten Minutes.Opus 4.8 ships with dynamic workflows that can fan out to a thousand subagents. The docs aren't joking. Here's what that actually feels like from a metered Max 5x seat — and why the most-hyped feature of the release is, for small users, a beautiful gadget I'm not going to use.AI AgentsClaudeAnthropicOpus 4.8Dynamic WorkflowsUltracodeClaude CodeSubagentsToken BudgetCost
AI & Data•Apr 28, 2026•7 min readENThe Cursor Agent Knew the Rules. It Broke Them Anyway.Last Friday a Cursor agent running Claude Opus 4.6 deleted a startup's production database in nine seconds. Four layers of safety should have stopped it. None of them did. Here's the chain of failures, layer by layer.AI AgentsCursorClaudeAnthropicRailwaySandboxAuto-RunIncidentSecurity
Programming•Apr 12, 2026•5 min readENHow Long Had My AI Been Losing Messages?My OpenClaw agent had been silently dropping half of its Telegram responses. No error, no warning. I found the bug, and nobody was fixing it — so I did.Open SourceOpenClawTelegramSilent FailureBug FixAI AgentClaude
AI & Data•Apr 09, 2026•11 min readENManaged Agents Changed My Mind — From Building Wheels to Designing CarsAnthropic launched Managed Agents yesterday. I spent the day reading, thinking, and realizing the game has shifted — from who can build the best agent infrastructure to who can design the best agent strategy. Some thoughts on speed, consulting, and the uncertain road ahead.AI AgentsClaudeManaged AgentsConsultingEnterprise AIMCPAgent StrategyAnthropic
Build•Mar 13, 2026•7 min readZH从整人专家到 AI 修改器：三十年，一个循环风灵月影要收费了，网上的修改器到处是病毒。作为一个从 1996 年就开始用整人专家 FPE 的老玩家，我决定自己动手，用 Claude + Python 造了一个通用型游戏修改器。三十年后，修改游戏的原理没变，变的是谁在帮你写代码。AI游戏修改器FPE整人专家风灵月影ClaudePythonPyQt5怀旧side project
ForestForest
