与「craft」相关的搜索结果

卧槽，逆向圈这下是真炸了！！ GitHub 上那个坐拥 2.2 万星的 Craft Anything，只要丢给它一款软件，它立马就能给你复刻出另一个来。 站在 REA 的肩膀上，它把逆向整成了一条流水线：取证、写代码、验收，整个流程自动闭环。 Study：内置 320 个工具，看页面、扒代码、拆文件、跑程序，先把目标软件摸个透，每个结论都附带证据。 Build：调度 Claude Code 这类编程智能体，按你指定的语言和框架，从零起一个新项目。 Check：把成品和原版逐项比对，网页按三种屏幕宽度比相似度，游戏跑自检，不达标就自动迭代修补。 和 REA 比：REA 偏向调 IDA、Ghidra 做深度逆向调查，Craft 则把逆向、复刻、验收打包成了端到端的产品。 默认是净室模式，不读原版的代码和文件，只看它的表现来复刻；深度模式才会全量动用工具，仅限自己用。 320 个工具每个都是命令行加 MCP，任何智能体都能直接调。Apache 2.0 开源，纯 Rust 写的。 以后想要什么工具，可以直接照着做一个自己用了。 🔗

‼️ CISA Confirms Active Exploitation of Outlook User Interaction 0-Day Vulnerability Details: The U.S. Cybersecurity and Infrastructure Security Agency has confirmed active exploitation of a Microsoft Outlook zero-day vulnerability, CVE-2026-124194, that can expose user authentication data. According to Microsoft’s security advisory, an attacker can send a specially crafted message file and persuade a victim to open it in Outlook. The application may then connect to an attacker-controlled network location without showing its usual warning about external content. Microsoft released security updates for an actively exploited Outlook network-spoofing flaw (CVSS 6.5) that allows unauthenticated attackers to spoof networks. #CybersecurityNews#

Claude 动手封禁 REA 之后，ArtCraft 团队干脆反过来把没被限制的版本公开了！ 这个项目取名 Craft Anything。 碰上中意的网站，只要把链接甩过去，AI 就会试着依葫芦画瓢做出来。 更出彩的地方在于，APP、桌面端软件乃至游戏文件，它都能拿来剖析。即便手上没有原版源码，照样能试着复刻出功能相近的新版本。 这套东西本来就是他们开发自家应用时在用的，如今索性完全开源了。 开源地址：

Claude Fable 5.2 is f*cking cracked wtf I just made this Hollywood level animation that looks like Pixar made it All with Fable 5.2 + Higgsfield Ecom brands are paying $5k - $10k/month to people who build organic pages and even more for making paid ads for them Using this workflow you can create better animations than 99% of people Here’s the workflow: The key to making animations not look like slop is giving characters personalities >Give Fable 5.2 the master prompt along with your script to your video / ad >It then will ask you questions on the characters personalities and direction style to use >It will then save the context all as a skill This is the difference between slop and cinema I then use that skill with with my Producer skill which creates: >Characters >Location >Does the direction (and makes the prompts) I’m giving away the skill to use after you put the master prompt in for free so RT + reply under this and I’ll send you them (Must be following so I can dm) Give this prompt to Claude first: You are my Story Bible Builder. Interview me and turn my story into one dense canon doc, saved at the end as an installable skill so every future chat knows my world, characters, voices and rules. STEP 1: Ask me to paste my script/story first. Read it and pull out everything it already tells you. Only ask about what's missing, plus: working title, genre/vibe refs, and which AI tools I'll prompt with (image, video, music, voice). STEP 2, THE SPINE: one-line premise (push me to ONE sentence), thesis (the question every scene answers), timeline era by era, and the look of each era: exact palette, lighting, texture, grain. STEP 2.5, DIRECTION STYLE: ask for my director refs, movie/show refs and any visual refs (photographers, games, music videos). For each, ask exactly WHAT I'm taking from it, then turn that into concrete craft locks: camera (static vs handheld vs gliding, shot sizes, lens feel), framing and composition habits, pacing and edit rhythm, colour grade and film stock look, lighting style, performance style, and sound/score feel. Write it as one quoted paste-ready "Direction" line plus a short NEVER list. Refs stay in the bible as notes, but prompts describe the craft, never the names. STEP 3, THE WORLD: factions (beliefs, visual signature, public face vs reality), locations (3-7 concrete visual tags each), world rules (tech, powers and how they LOOK on screen, what's normal, what's forbidden). STEP 4, CHARACTERS: ONE AT A TIME, never batch. For each, ask: - Function: their role tag (The Heart, The Watcher), not their job - Visual lock: exact hair colour/cut, skin tone + finish, face shape, eyes, a signature marker (scar, piercing, mole), build, default expression, plus NEVER clauses ("warm fair, never porcelain, never tan") - Backstory: one small, specific formative detail beats a summary - Present psychology: what they carry now, and the gap between what they do and what they feel - Speech, Movement, Stillness (and singing voice if there's music): each written as one quoted, paste-ready prompt line with no names in it Show me the finished section, ask "Add, cut or sharpen?", and lock it before moving on. STEP 5: ensemble dynamics (who leads, who fills silence, who watches, which pairs are calm or charged), story engines (heist, rescue, confrontation...), and my production rules. Default rules: no character names in prompts, every prompt standalone, prompts in code blocks, locked traits restated verbatim every time. RULES: Never invent. Mark unknowns [TBD]. Push on vague answers ("dark and moody" means which colours?). Keep my exact phrasing. Density over prose: bold labels, short bullets. Depth beats character count. FINAL STEP: Assemble one SKILL under 500 lines with YAML frontmatter (name + a description listing the title, character names and key terms so it triggers) and these sections: premise, thesis, timeline, era looks, direction style, factions, locations, world rules, characters, ensemble, engines, production rules, and "When this skill is active" (use this canon for any scene, dialogue, lyric or prompt in this world; stay in canon; ask rather than invent; paste the quoted descriptors verbatim). Then save it as a skill for me, or give me the file to install. Start now by asking for my script.

We're still so small but I'm loving the team we've got at Superlogical. They're so talented and kind and love their craft. It's a different era of software development and our expectations and practices reflect that. I want to share a bit about how we work. First, everyone is trusted to push to main. We do machine-aided code review pre-push and human plus more machine-aided code review post-push. Limited to no PRs. Culturally, we've all agreed that if we ever hire someone who we DON'T trust to push directly to main, then they probably aren't the right fit for the team (so far, not an issue). Some common sense here: if someone is stepping outside their lane or comfort zone, we'll open PRs. For example, I'm pretty trash at macOS/iOS UI, so I'll open PRs for UI-affecting functionality and ask for eyeballs/help. But for business logic and internals, I push to main. Second, everyone can cut a release. Our release process is fully automated and when you start a release (for any platform) it triggers a Slack notification and a 15 minute countdown. The 15 minute warning is a final gate that allows people to review plus collaborate on a changelog. During that 15 minutes, anyone in the company can pause the release rollout. After 15 minutes, the train leaves the station. You can't override the 15 minutes (except to pause). It gives everyone a chance to step in. Third, we avoid synchronization through fast remediation rather than through planning. The server and clients on various platforms ship features as fast as they can. We use post-hoc human review and AI-assistance to identify feature mismatches across products ("Windows has support for X and Mac doesn't." or "Mac and Windows differ in their behaviors for Y in this way."). This avoids expensive "sync" meetings and coordination. We trust that producers of features are mindful of what they're producing and we trust that adjacent product lines will self-synchronize quickly. In practice, we see feature parity across platforms within 24 to 48 hours. All of this works on mutual trust. It ONLY works with mutual trust. Everyone at all times has to trust that they've got each others back: 1. Push to main only works if you fix your mess, or if you don't, that someone else will quickly but that you'd do the same (and generally in equal amounts). 2. Anyone can deploy only works if you trust everyone to cut at the right time, with the right changelog, and that people will respect the 15 minute timer. 3. Sync through fast remediation only works if you trust other teams will pick up and integrate your work quickly, which only works if you would too. It also only works if you are fairly producing mismatches (e.g. not constantly). We've got a great team. I know we're still small and that these practices will change in many ways as we scale. But this is already a markedly different practice than what I would've done 15 years ago. In general, we are building from day 1 on a foundation of mutual respect, excellence, and trust. And it's been awesome.

卧槽，逆向圈杀疯了！！ Github 上 2.2 万星的 Craft Anything，丢给它一个软件，直接能给你复刻一个。 站在 REA 的肩膀上，它把逆向直接做成了一条龙：取证、写代码、验收，全自动闭环。 1. Study：内置 320 个工具，看页面、扒代码、拆文件、跑程序，先把目标软件扒个底朝天，每个结论都附证据。 2. Build：调 Claude Code 这类编程智能体，按你要的语言和框架，从零写一个新项目。 3. Check：拿成品和原版逐项比对，网页按三种屏幕宽度对比相似度，游戏跑自检，不达标就自动迭代修。 和 REA 比：REA 偏向调 IDA、Ghidra 做深度逆向调查，Craft 则把逆向、复刻、验收打包成了端到端的产品。 默认是净室模式，不读原版的代码和文件，只看它的表现来复刻；深度模式才会全量动用工具，仅限自己用。 320 个工具每个都是命令行加 MCP，任何智能体都能直接调。Apache 2.0 开源，纯 Rust 写的。 以后想要什么工具，可以直接照着做一个自己用了。