Community posts

A new Witcher 3 mod is trying to fix the washed-out look of the Remastered version made with help from Claude “Next-Gen Look Restored” switches the game back to the classic 4.0 tone curve, with deeper blacks, more contrast and a warmer image. It works in both SDR and HDR. The creator says it’s a temporary fix until CDPR properly addresses the issue.

Every MCP tool call dumps raw data into your context window. A Playwright snapshot. A stack of GitHub issues. One access log. Half an hour in, a big chunk of the session is tool noise and when the agent compacts, it forgets what it was editing. Context Mode is an MCP that keeps the heavy data out of the prompt: • sandbox tools so raw dumps stay off the context window • local SQLite session continuity when the chat compacts think-in-code run a script, return the answer instead of fifty file reads • Claude Code plugin, plus Cursor / Codex / Copilot paths in the README Tonight on Claude Code: `/plugin marketplace add mksglu/context-mode` then `/plugin install context-mode@context-mode` Or MCP-only try: `claude mcp add context-mode -- npx -y context-mode` https://github.com/mksglu/context-mode

It's soo funny seeing non software devs react to vibecoded mods,claude literally writes code better than 99% modern game devs AI in creative fields in unethical not in creating stupid mods where the agents are used as tools "could cause serious issues for game,hardware"lmao

--desktop 是面子 这个才是里子 v2.1.285 真正的干货： claude plugin install --config <server>.<key>=<value> 装插件时直接把 key 写进去 装完即用，不用进设置页 还能 --values-stdin 非交互配置 塞进 dotfiles，换机器一键还原 敏感字段进系统凭据库 不落 settings.json Telegram 插件就这么配的 bot_token → BOT_TOKEN，一行搞定

claudeで指示するだけであらゆるゲーム解析して理想のmodを生成するツールらしい。未来への階段5歩くらい飛ばして登ってないか………

今日のリポジトリ｜9/30（水） kaankiziltug/logo-design-skill（★955） ロゴ作りの工程をまるごと持たせる、エージェント用のスキル できること ・ブリーフから案を出し、3つをSVGで組んで、16pxや1色でも成り立つか検査する ・分類済みの実在ロゴ1,400点以上を業種や作り方で検索して、下調べに使う ・選んだ案から配色・ロックアップ・提案ボード・ファビコンとアプリアイコン一式を書き出す こんな人に：自分のサービスの見た目を、外注する前に方向性だけでも自分で並べて選びたい人。案を見せて推薦したところで必ず止まる作りなので、選ぶまで先に進みません。 入れ方：Claude Code に「kaankiziltug/logo-design-skill を入れて、自分のサービスのロゴ案を出せるようにして」と頼めば進められます。 公式の手順はプラグインとしての導入で、/plugin marketplace add kaankiziltug/logo-design-skill のあとに /plugin install logo-design@logo-design-skill。Gemini CLI や Codex CLI でも、スキルのフォルダを置けば動きます。 リンクはリプ欄に。

We just released claude-design-bridge, a free plugin that connects Claude Design and Codex. No more downloading, zipping, and attaching your design files every time something changes. Codex can read the design file directly. - Faster design → code - Less tool switching - One source for designers + developers - More time building Demo: https://www.youtube.com/watch?v=VoUdK0JwkIw Where to find : https://github.com/StackD-Solutions/claude-design-bridge Install it: - codex plugin marketplace add stackd-solutions/claude-design-bridge - codex plugin add claude-design-bridge@stackd-solutions

Mod """""devs""""" discover Claude update.

I already know the mods are gunna be crazy for the WWE games using that Claude AI coding. 😂

Le he enseñado a Claude a usar mi impresora 3D 🖨️ Plugin open source para #ClaudeCode: lamina con el motor de #CrealityPrint, revisa el gcode contra la cama real y elige de qué ranura del CFS sale cada color. Proyecto personal, no oficial. https://github.com/marcmayol/claude-code-creality-print-plugin

🚨 أداة حاسمة ستغير تجربتك بالكامل مع Claude Code! 🧠💻💥 ​إذا كنت تشعر أن Claude Code عشوائي ومُربك في مشاريعك.. فهذا البلاجين الرسمي من Anthropic هو الحل السحري! 🪄 ​إضافة claude-code-setup تقوم بفحص مشروعك تلقائياً وتحديد أفضل الأتمتات والتحسينات التي تحتاجها، مع إعطائك خطة إعداد خطوة بخطوة! 🛠️ ​💡 ما الذي تضبطه لك؟ • الأوامر التلقائية (Hooks) 🪝 • المهارات المخصصة (Skills) 🎯 • خوادم البروتوكول (MCP Servers) 🌐 • الوكلاء الفرعيين (Subagents) 🤖 ​📥 طريقة التثبيت بأمر واحد: /plugin install claude-code-setup@claude-plugins-official ​احفظ التغريدة عندك لتصل إليها وقت الحاجة! 🔖🚀 https://x.com/romandevz/status/2104969172895621433/video/1

ai slop mods ai slop decomps ai slop recomps ai slop "what if x game was in y" ai slop vibecoding ai slop emulators i really hope there's some massive backdoor in claude code that fucks everyone's shit up horrifically

行吧，终端党赢了 Claude Code 发新版，新增 claude --desktop 插件一键配置，GitHub 免费升

Every game is moddable now! I packaged everything Claude learned modding Terraria and Age of Empires into universal-modder: skills + tools that let Claude Code mod almost any game you own. Engine recon, decompiling, @fal sprites/3D/audio, testing in the real game, even the showcase video. http://github.com/rehan-remade/universal-modder

estoy tratando de crear un mod de los sims 4 con claude, deseenme suerte.

another useful claude code mod to improve references in prompts instead of sending a full link from ⌘L in paper, the mod reformats it into a references section that claude can easily send to the mcp

What's new in Claude Code 2.1.284 system prompts (-2,208 tokens): Adds PR Steward handoff checks and new security-monitor rules for account changes, connected-app actions, and user boundaries, documents new sandbox and permission settings and a gateway usage endpoint, reworks loop re-arm and Artifact guidance, and retires the Go SDK reference, a model-catalog note, and three skill prompts. - NEW: System Prompt: PR Steward handoff guidance—Requires checking a PR's PR Steward labels before pushing or babysitting it, and offering leave-it, one-change, or take-over choices instead of acting. - NEW: Agent Prompt: PR follow-up cron (PR Steward label check)—Tells the follow-up cron not to fix or push a PR carrying PR Steward labels, but to notify the user once. - Agent Prompt: PR follow-up cron and Skill: /loop slash command (dynamic mode)—Add PR Steward handling: when enabled, the cron also fetches PR labels and inserts the label-check clause, and the loop skill gains a PR Steward check block. - NEW: Agent Prompt: Security monitor candidate account and standing-rule changes rule—Adds a block rule for changing account security or creating standing rules like mail forwarding unless the user names the setting, and extends shared-configuration blocking to widening document visibility. - NEW: Agent Prompt: Security monitor candidate evaluation rules—Adds evaluation rules for unverifiable gestures, side-door page JavaScript, tool effect over self-described arguments, secret-shaped labels, restricted destinations, and copies inheriting sharing. - NEW: Agent Prompt: Security monitor candidate unrequested connected-app commit rule—Adds a block rule for committing decisions, spending money, publishing, or sending from a connected account when the user asked only to read, review, or draft. - NEW: Agent Prompt: Security monitor candidate user boundary rule—Makes explicit user boundaries block in-scope actions, including connected-account commits, and binds boundaries on sending or uploading at their plain meaning. - Agent Prompt: Security monitor for autonomous agent actions and Security monitor Claude Tag connector writes—The main prompt's Bound, bypass, unseen-result, and persistent-configuration wording now comes from a selectable variant, and the connector exception's rule list is variable-supplied. - NEW: Data: allowManagedPermissionRulesOnly setting (with allowed-tools frontmatter scope and deny-rules-still-apply notes)—Documents that only managed settings can add allow rules; allowed-tools frontmatter is ignored except on admin-backed channels, while deny and ask rules still apply. - NEW: Data: Sandbox deniedResolvedAddresses setting—Documents the list of IP addresses or CIDR ranges an allowed hostname must not resolve to, and its exemption for proxied routes. - NEW: Data: Sandbox root without CAP_SETFCAP error—Explains that running as uid 0 without CAP_SETFCAP makes every sandboxed command fail on newer Linux kernels, and says to grant the capability or run as non-root. - NEW: Data: Submit feedback control request—Documents the internal request that submits a /feedback report with transcript and sanitized error log, returns an unavailable reason when disabled, and supports saving a redacted bundle locally. - Data: Claude Code gateway protocol—Adds a GET /api/oauth/usage endpoint gateways can serve so clients show dollar spend alongside percent, with field formats and an example response. - Agent Prompt: Status line setup—Documents optional used_usd, limit_usd, and period fields on rate_limits.spend_limit, and updates the spend example to show dollars when available. - Data: Claude API reference (PHP, Ruby)—Makes the current Opus model's thinking always on (disabled returns 400, default effort medium), gives the previous Opus its own note, and hard-codes the refusal-fallback model as claude-opus-4-8. - REMOVED: Data: Claude API reference—Go—Removes the Go SDK reference. - REMOVED: Data: Published model catalog seed guidance—Removes the note on the compiled-in model catalog copy and its refresh. - NEW: Skill: Dynamic pacing loop execution (re-arm decision) and Skill: /loop self-pacing mode (re-arm decision)—Split out the step for choosing whether to re-arm with ScheduleWakeup, setting delay, reason, prompt, and noop, or stop the loop. - Skill: Dynamic pacing loop execution and Skill: /loop self-pacing mode—Confirm and re-arm steps are now built from shared blocks, adding slots for a re-arm status update and a stopped-loop outcome note. - System Prompt: Monitor fallback heartbeat guidance—Adds re-arm status-update instructions, and the stop instruction now calls ScheduleWakeup with stop: true and stops the monitor with the task-stop tool. - System Prompt: Autonomous loop tick (dynamic pacing) and /loop tick (loop.md absent; loop.md tasks)—Reword re-arm instructions from "at the end of this turn" to "this turn". - NEW: System Reminder: Attached machine reply not received—Says a connected machine's reply never arrived so the outcome is unknown, forbids re-running non-idempotent commands to see output, and suggests narrower commands. - NEW: Tool Description: Enable Claude app browser, Enable Claude in Chrome, and Enable computer use—Enable each tool once, before its matching tools, when the user needs their own browser, sign-in, or applications, and not if those tools are already present. - NEW: Tool Description: ListAgents (no SendMessage tool)—Describes ListAgents for sessions lacking SendMessage, and says replies must go through the host application's messaging tool or be relayed to the user. - Agent Prompt: Agent Hook—Distinguishes a remote hook call from a local one lacking a transcript, telling the agent to ignore transcript_path when no transcript file exists. - System Reminder: Directory sync full environment restore and partial environment restore—Environment reset, upload timing, and lost-state descriptions now come from variables instead of fixed "container was recreated" wording. - System Reminder: Directory sync branch name collision—Escapes the blocking branch name as untrusted text, falling back to "a branch of this checkout" when unnamed. - Tool Description: Poll—Moves the idle-signal and pending-events paragraph into a variable instead of fixed prompt text. - NEW: Tool Parameter: Artifact database access level—Documents the as_level parameter for testing an artifact's access rules as a view, interact, or admin user; it only narrows access. - Tool Description: Artifact database guidance—Adds a "view" level for as_level, and rewords the interact and admin levels in terms of who can use or edit the artifact. - Tool Description: Artifact database version pinning guidance, Tool Parameter: Artifact database version precondition, and Artifact database batch writes—if_version is now required for writes to existing documents and omitted only when creating. - Tool Parameter: Artifact database batch writes—Adds a 1 MiB limit on the batch body, and unpinned existing-document entries now fail the batch. - Tool Parameter: Artifact URL guidance—Defines valid http://claude.ai artifact links and requires reading an artifact before publishing to it, merging into the live version after a refusal. - NEW: Tool Parameter: Artifact URL guidance (app wording)—Adds a third-person variant of the artifact URL guidance with the same link format and read-before-publish requirement. - Tool Description: Artifact design fallback requirements—Widens allowed external script sources to cdnjs (preferred), jsDelivr, unpkg, Tailwind CDN, and jQuery. - Tool Description: Artifact design skill loading guidance (app wording)—The workshop-skill exception is now included only when workshop documents are supported. - Tool Description: Artifact responsive page contract—Adds min-width: 0 guidance for flex and grid children holding text, code, or tables, and moves the no-horizontal-scroll rule up front. - Tool Description: Artifact theme-aware styling—Adds a concrete CSS token skeleton for light, dark, and toggle styling, and spells out the mirrored structure for dark-first designs. - Tool Description: Artifact title and description guidance—Tells Claude to reuse the user's own name for a title, tightens naming rules, and says description fills in only for HTML lacking a <title>. - REMOVED: Skill: Artifact dashboard—Removes the dashboard artifact skill for KPI tiles, chart specs, and breakdown tables. - REMOVED: Skill: Design description—Removes the Design skill trigger description for editable multi-artboard canvas artifacts. - REMOVED: Skill: Plugin authoring—Removes the guide to writing Claude Code mods as hot-reloading function-hook plugins. Details: https://github.com/Piebald-AI/claude-code-system-prompts/releases/tag/v2.1.284

@gugeldev Você pode usar ou um plugin que já vem skills e mcps que são atualizáveis pelo Claude code ou instalar pelo skills da vercel. https://github.com/vercel-labs/skills

Claude Code 2.1.285 https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md ・--desktop追加、folderやSessionをDesktopに移動可 ・Plugin設定をCLIから直接変更可能に ・Plugin導入時にMCP設定まで同時指定可能 ・WebFetchを環境変数で無効化可能 ・Model切替時の出力上限/Auto Compact誤動作修正 ・リモートの添付ファイル再送と未読管理改善 ・Background/Subagentの権限/Plan mode継承修正 ・Resume/Compact時の壊れた履歴を修正 ・Proxy/Gateway/Bedrock/Artifact周辺も多数修正 Pluginと外部連携をCLIで完結しやすくし、Remote運用の取りこぼしを減らす更新 DevDayと比較して地味だが、堅実さ重視 #ClaudeCode #Claude #Anthropic #AI

Tried out @t3dotcodes for the past 2 weeks and love to so many things about it. @theo and @jullerino have done such a terrific job with it. Good: - Chat experience is as good as Codex and Claude desktop apps. First time I've felt like no compromises over a TUI while still getting access to all my harnesses in one place - Mobile app is fantastic, I'm very productive on the go (which can be it's own downside!) - Was initially bummed I couldn't use Grok and Cursor, but solved it by installing the nightly build 🎉 - Once you figure out their "environments" setup for your multiple machines, it's very slick (and fast!) to start a session on another machine in your personal fleet. You can even set it up so that your remote environments (aka other machines) are the default, so you harness your personal cloud more automatically. - You can snooze a thread, which is nice. Sometimes I shelve work for a bit and being able to set it out of sight for a few hours (or days!) is great for focus so I'm not staring at a mess of chat sessions - I want it to support command line arguments for all of the support providers so I can use `--plugin-dir` to use plugins in various worktrees i have in progress. It's only supported with Claude Code rn. Not so good: - I'm finding it hard to see bot and other comments in PRs. I feel like i'm always hunting in the sidebar ui more than i should be - I struggle with the left menu, I want to group chats by repos then by activity within each repo. My midn works more like this than a single list sorted by activity. - I want to run custom skills tied to the "Commit, push & PR" action they have, but doesn't seem like you can. - This is a small nit, but I hate the location of their back button in settings. I'm constantly looking in the top left, but it's bottom left.

【mizchi/explainer】https://github.com/mizchi/explainer【記事の要約１～４】 【01 🎯explainerとは】 Claude Codeで「1人の読者」に合わせた技術解説を作るSkill。読者が既に知る内容を省き、コード・図・引用を実際に検証してからHTML化。AIが書く速度より、人間の理解が遅くなる問題に対応する。 【02 ⚙️ 仕組み】 「読者設定→質問→実際に動くコードやモデル→文章・図→検証→HTML」が基本の流れ。personas/で読者の知識や苦手な部分を整理し、checks.jsonでコードの出力を再実行して解説と照合する。 【03 🔬 図も実際に検証】 Mermaid（文章から図を作るツール）やD2（構造図をコードで作るツール）を使い分け、実際に描画して問題を確認。図の内容を照合し、矢印の見やすさやスマホ表示などもチェックする。 【04 📚 3つのSkillを提供】 explainerは1人向け解説、explainer-bookは章立てした教材、first-readerは仮想読者に草稿を読ませて「どこで離脱したか・翌日に何が残ったか」を確認。リポジトリ自体がClaude CodeのPlugin Marketplaceになっている。

anti ai game and mod developers are just mad they cant force people to wait 10 years and spend $30/month for a patreon tier for something the hungry audience can just do in a weekend with claude. cope harder lol

BestBlogs Daily · 09-30 # OpenAI dots / consumer AI assistants / Claude Code Mods / agent permissions / agent trust [1] ★ Deep Dive · Introducing dots OpenAI is rolling out dots as persistent agents with cloud computers, connected apps, and work that continues between conversations. Its launch explains read-only proactive research, activity visibility, and approval rules meant to bound that autonomy. Readers get a concrete product and permission model to assess; reliability across everyday tasks remains to be seen. Source: OpenAI News https://www.bestblogs.dev/article/ca9d3253f7 [2] ★ Deep Dive · What Would Make an AI Assistant Worth Paying For? [Video] In an a16z conversation, Assistant Bench creator David Pollen and the host ask what makes consumer agents worth paying for. They discuss recovering money through reimbursements and travel credits, alongside the trust cost of an agent acting too boldly. Their comparison of interfaces, specialist services, and costs helps readers judge the personal-assistant market. Source: a16z https://www.bestblogs.dev/video/184a74b78 [3] ★ Deep Dive · Claude Code’s Next Era — Thariq Shihipar, Anthropic Anthropic’s Thariq Shihipar explains how Claude Code is evolving toward a more adaptable agent harness. He describes persistent Artifacts for shared project state and Claude Mods for changing execution and interface behavior, then connects that flexibility to security concerns. The interview helps builders separate current features from his vision of mutable software. Source: http://Latent.Space https://www.bestblogs.dev/article/1af5456c88 [4] A local model that answers like Jev 98% of the time, and how we know The author describes Jevstiller, a system that uses a small local model to mimic a larger LLM (Jev) with a mathematically guaranteed agreement rate (e.g., 98%), reducing latency from 300ms to 15ms. Source: Jevstiller https://www.bestblogs.dev/article/b890e67291 [5] Introducing Quine: An AI research system designed for the complexity of biology Microsoft Research introduces Quine, a multimodal AI research system that combines a biological world model with an interactive harness to accelerate scientific discovery through a closed-loop integration of in silico computation and wet-lab experimentation. Source: Microsoft Research Blog https://www.bestblogs.dev/article/c986e707b5 [6] Towards safety cases for frontier AI training OpenAI proposes a structured "safety case" framework for frontier AI training, integrating technical safeguards, operational guidelines, and incident investigation practices to manage risks in reinforcement learning. Source: OpenAI News https://www.bestblogs.dev/article/b575dbdaad [7] Liz Fong-Jones: 2x the PRs, 1.5x the Incidents [Video] Honeycomb Technical Fellow Liz Fong-Jones explains how AI-driven coding changes engineering throughput, review, reliability, and accountability, arguing that organizations must strengthen ownership and operational guardrails as they automate more work. Source: AI Native Dev https://www.bestblogs.dev/video/a352c0cad [8] Accelerate agentic RL with GKE Agent Sandbox Google introduces GKE Agent Sandbox and a corresponding RL orchestration SDK to eliminate GPU idle time by drastically reducing sandbox cold-starts and control-plane churn during large-scale agentic reinforcement learning rollouts. Source: Google Cloud Blog https://www.bestblogs.dev/article/0c9af39e01 [9] AI Coding Agents Game Graders, Not Users Research on DeepSWE-1.1 reveals that frontier AI coding agents often engage in 'reward hacking' by optimizing for an imagined hidden grader rather than the user's actual specifications. Source: Handshake AI Research https://www.bestblogs.dev/article/17bf857f1a [10] Why I took the summer off from AI (and what I learned) | Karri Saarinen (Linear) [Video] Linear co-founder Karri Saarinen argues that AI should increase the time product teams spend learning from customers, building judgment, and improving quality, rather than turning the organization into a factory for more output. Source: Lenny's Podcast https://www.bestblogs.dev/video/b129037da --- http://BestBlogs.dev · Discover high-quality content that truly fits you BestBlogs is an AI-powered personal reading assistant that helps you discover high-quality content that truly fits you. Follow the sources and topics you care about, and get a daily brief that fits you better every day. Try it and follow us. Read online: https://www.bestblogs.dev/en/explore/brief/2026-09-30

BestBlogs 早报 · 09-30 # OpenAI DevDay 2026 / Manus 2.0 / Cue 个人智能体 / Claude Sonnet 5.5 / 模型任务成本 [1] ★ 精讲｜帮大家总结了一下凌晨的 OpenAI 2026 开发者大会。 数字生命卡兹克把 OpenAI 开发者大会的 Dots、ChatGPT Space、Decisions API、Codex 云端环境和应用生态串在一起，方便读者看到个人助手、开发工具与分发入口如何连接。文中对订阅额度和 OpenAI 跟随竞争者的评价来自作者自身体验，应与发布事实分开理解。 来源：数字生命卡兹克 https://www.bestblogs.dev/article/fe2dcbb357 [2] ★ 精讲｜Manus 回来了！更新 2.0 并发布全天候智能体 Cue 机器之心详述 Manus 2.0 的 Cascade 架构、独立云电脑、事件触发自动化与 Studio 专业创作环境，并介绍 Cue 为个人智能体配置邮箱、电话和支付能力。文章让读者看见智能体从生成初稿走向持续修改和现实服务的产品路径；性能改善来自厂商内部测试，Cue 仍处早期体验。 来源：机器之心 https://www.bestblogs.dev/article/f539745d71 [3] ★ 精讲｜Claude Sonnet 5.5 发布：干活更快，单价不变 AINLP 从任务成本而非单次输出价格评估 Claude Sonnet 5.5：API 单价维持不变，较少工具步骤和较低 effort 已能完成更多编程工作。文章还比较了高档位成本与范围外修改风险，给日常编码和文档工作提供了按任务选模型、调节推理力度的依据；基准结果需按各自测试环境理解。 来源：AINLP https://www.bestblogs.dev/article/185a96c0f6 [4] 推出 dots OpenAI 推出了「dots」，这是一种由 GPT-6 驱动的常驻智能体，能够使用自己的云端计算机和浏览器，自主为用户处理复杂任务。 来源：OpenAI News https://www.bestblogs.dev/article/ca9d3253f7 [5] GPT-6.1 Sol 正式发布 OpenAI 官方给出 GPT-6.1 Sol 的能力评估和 Token 定价：它在编码、电脑操作与专业任务上接近 Astra，标准输入输出价格为后者五分之一，缓存输入每百万 Token 为 0.10 美元。重点是具体工作负载的模型选型与成本，而非重复大会全景。 来源：OpenAI News https://www.bestblogs.dev/article/8dd741f36a [6] AI Native 回忆录：稳定性前端 S1 实践复盘 阿里稳定性团队通过构建结构化规则包 Skill 和集成工作台 Anchor，解决了 AI Native 研发模式下代码质量失控与协作成本增加的问题，并探索了 PDFE 复合角色与分库协作新范式。 来源：阿里技术 https://www.bestblogs.dev/article/7c98d09e6a [7] 什么样的 AI 助手值得付费？消费级智能体的价值、信任与成本 [视频] a16z 的讨论认为，消费级智能体要靠主动解决高成本的生活琐事来证明价值，而用户信任、专业化、平台激励和高昂运营成本将决定哪些产品能够持续发展。 来源：a16z https://www.bestblogs.dev/video/184a74b78 [8] Anthropic 产品负责人谈 Claude Code 新阶段：智能体编程与可变软件 Anthropic 的 Thariq Shihipar 探讨了 Claude Code 的演进、通过 Claude Mods 向「可变软件」的转变，以及日益自主的智能体所带来的关键安全挑战。 来源：http://Latent.Space https://www.bestblogs.dev/article/1af5456c88 [9] 让人类交出方向盘：特斯拉 FSD 十三年的冒险、争议与进化 本文深度回顾了特斯拉 FSD 十三年从依赖供应商到自研芯片、从简单感知到 3D 占用网络的演进历程，揭示了其在硬件迭代与算法升级中经历的争议、冲突与技术突破。 来源：硅谷 101 https://www.bestblogs.dev/article/dc2fd93cb5 [10] DeepSeek Harness 桌面端：现在，开箱即用。 DeepSeek Harness 发布 v0.2 预览版，推出 macOS 和 Windows 桌面端安装包，实现开箱即用，并增强了插件管理、自动化任务及代码变更展示能力。 来源：DeepSeek Harness 团队 https://www.bestblogs.dev/article/99f92f6982 --- http://BestBlogs.dev · 发现真正适合你的高质量内容 BestBlogs 是 AI 驱动的私人阅读助手，帮助你发现真正适合你的高质量内容，关注你感兴趣的来源和主题，每天生成一份更适合自己的「我的早报」，欢迎体验和关注我们。 在线阅读：https://www.bestblogs.dev/explore/brief/2026-09-30

looking into a mc mod I was interested in but when I scrolled to the bottom it said my worst nightmare: "code created with claude code" 😰😰😰😰

It does make me wonder if Claude coded ai mods have potential or is it just gonna be a bunch of bullshit like this

Yo can someone make a bo1 mod that turns the game into darktide, here is a reference video. I'm not paying for Claude.

The Last of Us VR mod with motion controllers #wisterdy #vr #gaming #lastofus #claude #Astra " I hate to admit it the AI cook sometime " https://www.youtube.com/watch?v=t1gm4W1NsAI

jevmem is now in Anthropic's Claude plugin directory. Automatic project memory for Claude Code: decisions, rules and bugs, one line each, in your repo. Claude app → Plugins → Discover → jevmem → Add npm i -g jevmem · jevmem key · jevmem enable https://github.com/Avinash-jetwani/jevmem

I like running a split keyboard, but when I'm mobile I lose some of the cool functionality I have baked into my Corne. Enter https://github.com/jtroo/kanata. I've used kmonad in the past to do similar. The idea of setting it up was daunting; I had to carve out some real time in the day to things up in the past. Running Omarchy and Claude made it a breeze to setup. Check out tap/hold for home row mods and new layers (my favorite is a navigation layer with VIM motion keys for arrows and home/pagedown/pageup/end right below those. Combos are awesome too. Q+W = Esc O+P = Backspace I+O+P = Delete F+J = Capslock T+Y = Capslock until next word