bro to bro, set Opus 5.5 to medium

摸索了几个月，我目前的视频工作流是： 1. Chatcut 2.剪映专业版 3.MiniMax H3 居家过日子必备 我读书多，不会骗你们的 🙂

十天时间，个人手搓300多个分镜，消耗十万积分！ 做出的“爱死机”克苏鲁科幻片《Onyx》 本视频80%由白模辅助生成 近未来AI突然自己进化出了血肉之驱 并且除了可以蒸馏人类的知识，甚至还开始“蒸馏”人类的肉身 它继承的越来越多，我们存在的理由却越来越少 当机械的心脏长出血肉 最后，人类成了新生命的养料

🙂呵呵，那时候还真是狂妄 ，年轻不懂事 Too young, too simple, sometimes naive

我现在手握 5 个 Google AI Pro 账号 从来没打过这么富裕的仗 🤣🤣哈哈哈

在我的电脑上 登录着两个Claude账号 一个是美区，没有封号， 另一个是尼区，也没有封号。 ------ 鲁迅 Day 5

🎉《GPT越狱记》 --- 欧阳修 极客之意不在算力，在乎图影之间也。 生图之乐，得之心而寓之显卡也。 提示词👇🏻

🚨 MAJOR ANNOUNCEMENT: YouTube will start prioritizing original content and reducing re-uploads (RIP clipping) YouTube has just announced one of the most impactful changes to its short-form algorithm. Its recommendation systems (the algorithm) will start to further prioritize original content and reduce the reach of content re-uploaded from other creators without adding anything of your own. They highlight this shift toward originality, where channels with primarily re-uploaded content will see less distribution within the Shorts feed. It also seems like they’re really trying to put the focus on content with real value, while highlighting that YouTube doesn’t want as many “minor technical edits” or “template-based” changes. Which is basically all that clipping was in this industry. Overall, while this will piss off some people whose entire careers were built on the premise of other people’s content (mostly clippers), the change itself is amazing and will finally address the overall quality issue we’ve been seeing.

借图用H3本地生成视频，还不错。 H3 提示语： subject_definitions: <Subject 1>: The single referenced young adult fashion model: slim build, light skin, long dark brown hair, soft oval face, dark eyes, gentle expression, small light dangling earrings, and a pale blush-pink sheer long-sleeved floral dress with fitted waist and layered ruffled skirt; barefoot throughout. Source(s): <Picture 1>, <Picture 2>. summary: A continuous vertical intimate fashion-photo-session video: the referenced woman moves fluidly through the nine studio poses in the supplied storyboard, beginning from the supplied first-frame kneeling pose. Hard constraints: Exactly one continuous 15.083333-second vertical shot; no cuts, hidden transitions, narration, subtitles, or spoken dialogue. | Open on the rear-facing kneeling, over-the-shoulder composition established by <Picture 1>. | Retain the same single young adult woman, long dark hair, pale-pink floral sheer ruffled dress, bare feet, warm blush seamless studio, and soft fashion-portrait lighting. | Follow the nine-pose reference progression from <Picture 2> through continuous body repositioning; do not depict the reference as a visible collage or introduce additional people, furniture, text, or props. | Keep face, hands, feet, and dress inside the central vertical safe area during reframing. Generation controls (authoritative) — Prompt-text rule: every section label and descriptive sentence is silent production direction, never speech content. Do not speak, recite, chant, sing, or read this prompt aloud. Background music: enabled. Generate a context-appropriate non-diegetic score unless the user explicitly requests silence or no music. Dialogue permission is enabled, but this plan contains no dialogue lines. Generate no character speech, conversation, chanting, singing, or other vocals. Dialogue language: automatic. Preserve each explicit <d> language tag; otherwise use the language required by the scene and user intent. Narration: disabled. Generate no narrator, voice-over, spoken description, or audible reading of prompt text, section labels, instructions, asset tags, or metadata. Subtitles: disabled. Do not render subtitles or dialogue captions. Creative direction (authoritative within explicit user intent and reference fidelity) Priority: Apply these creative directives clearly wherever they do not conflict with explicit user requirements, identity retention, or fixed first/last-frame composition. Camera execution: Follow the action continuously in a single unbroken take, using only motivated pans, small translations, and reframing needed to preserve spatial clarity. Look: Treat connected visual references as authoritative for identity, wardrobe, objects, environment, composition, lighting, color, and medium; do not impose an unrelated restyle. Pacing and composition: Use moderate, readable motion with natural acceleration and enough settling time for each important beat. Use exactly one continuous shot with no hidden or explicit cuts; express all action beats through blocking, performance, camera motion, reframing, and synchronized sound. An explicit user-requested shot or cut count remains authoritative over this density preference. Compose for a vertical frame: protect faces, hands, and essential action inside the central mobile-safe area; favor depth movement and avoid losing subjects at narrow side edges. retention_analysis: <Picture 1> -> first_frame: Use <Picture 1> as the fixed first-frame reference and primary authority for the woman’s identity, rear-facing kneeling pose, over-the-right-shoulder gaze, pale-pink floral dress, barefoot styling, warm peach-beige seamless studio, and soft directional fashion lighting.. <Picture 2> -> composition: Use <Picture 2> as the strict pose-storyboard reference for a continuous sequence of seated, kneeling, hands-and-knees, reclining, and floor poses; retain its vertical blush studio fashion-editorial presentation, wardrobe, subject appearance, and bare-foot styling.. detailed_description: [Shot 1] At 00:00.000-00:15.083, In a quiet blush-pink seamless studio during an intimate fashion photo session, <Subject 1> begins exactly in the rear-facing kneeling pose: folded legs beneath her, bare feet toward the camera, torso turned back over her right shoulder with a calm closed-lip smile. Her long dark hair, pearl-like earring, and pale blush floral ruffled dress are clearly visible under soft warm editorial light. She then transitions fluidly through the supplied storyboard’s pose language without cuts: she turns into a seated pose with one leg extended; rises briefly to hands and knees and looks toward the lens; lowers into a stomach recline with her head supported by one hand and one bent leg lifting behind; pivots to lean back on one hand with one knee upright; returns to a centered kneel with hands together on the floor; sits with an arm extended overhead; draws her knees close in a compact seated pose; and finally reclines onto her back, hair spreading across the floor as her crossed ankles lift elegantly. Each pose gets a brief settle-and-hold beat suitable for shutter capture. Her performance stays poised, gentle, and natural; no vocal performance. Camera: One uninterrupted stabilized handheld fashion-camera take at low-to-mid height. Start in a full-body rear three-quarter framing matching the first image, with her feet in the lower foreground and her turned face in the upper central safe area. Make a slow, motivated leftward arc and slight push-in as she rotates into seated poses, then gently pan and make small lateral translations to retain her whole body through kneeling, floor, and reclining positions. Ease backward and slightly higher for the final lying pose so raised crossed legs, face, and spread hair remain fully visible. Natural parallax only; no optical zoom, whip pan, jump cut, or collage framing. Motion: Moderate, readable continuous movement: initial stillness and over-shoulder hold, then smooth pivots, controlled lowering and rising, and carefully paced floor transitions. Hair and ruffled skirt respond softly to each movement. Every pose briefly stabilizes before the next transition; the final raised-leg recline settles calmly. Synchronized audio: 0.0-15.083333: low-volume non-vocal fashion-editorial instrumental score.; 0.0-2.0: soft fabric movement and a restrained camera shutter click as the opening pose settles.; 2.0-12.8: light floor contact, dress rustle, and occasional quiet shutter clicks aligned with each completed pose.; 12.8-15.083333: fabric settles; final subtle shutter click and sustained soft studio ambience. overall_soundscape: Quiet studio atmosphere with soft fabric rustle, subtle bare-foot and hand contact on the seamless floor, faint breathing during exertion, and discreet camera-shutter clicks timed to settled poses. No dialogue, narration, captions, crowd noise, or other voices. No narration, voice-over, spoken description, or audible prompt reading. non_diegetic_music: A soft, understated airy fashion-editorial instrumental bed plays throughout at low volume, with gentle warm synth pads and sparse muted percussion; no vocals, lyrics, or speech.

🎉 Gemini 4 蛰伏许久，一出手就是王炸！ 写作领域 直接霸榜 在中文、英文等多语种文本生成遥遥领先 曾经的小甜甜 Opus 4.6 已成牛夫人 迫不及待想使用了 以上🙂

🎉牛皮！刚测了 Cloudflare 新出的 cf CLI，直接把整个控制台 3000+ API 搬进了终端！ 以后从买域名、配解析到写 Worker、绑 WAF，在终端一条龙搞定！ 核心亮点： 1. 自然语言语义搜索：降维打击传统的 --help 内置了 cf cli search "<自然语言>"，用大白话提问就能返回最匹配的命令和说明 ⚠️ 目前不支持中文 ----- exp : cf cli search "search domain" ----- 2. 🔥终端查域名&购买 （见录屏） cf registrar registrations check <要查询的域名> 秒出可用状态和成本价，甚至能直接走终端交互表单扣款购买 3. 全链路终端闭环：以后配合 Claude Code / Cursor，从买域名、配解析到写 Worker、绑 WAF，一条命令全在终端闭环