Community posts

Running several coding agents in @herdrdev ? Termaxa has a Herdr plugin now: one action starts Claude Code under the gate, a refused rm -rf shows on the sidebar as "termaxa deny", and the record opens beside the agent it stopped. herdr plugin install termaxa/termaxa

You Don't Want Me-Cool. *new music release* For Adult Audiences Only. "this is my unofficial official comment" "you didn't want me, so now you know what you're missing, ok" "i don't like ugly anti-break-ups, so this is a way of avoiding an online feud" My Lyrics: (in case you were wondering) don't kneel before zod don't run after claude you like my new mod don't be an old fraud i can't be bought you can't be taught now we're gone i'm just a pawn zaptica africa breaking bad crying mad it's so funny where's my money apology love is for free

anthropic has been cooking so hard lately that my favorite openai devday announcement is that codex can now use compatible claude code plugins codex plugin marketplace add owner/repo what a time to be alive

a staggering amount of mods to make games into VR releasing right when claude improves

Performative mod enjoyer: Hold on let me wait and see if Claude made this game mod before I decide if I can enjoy it or not so I'm not taking jobs away from mod authors o algo

Claude Code es un lío. Hasta que instalas esto. Hay un plugin oficial de Anthropic llamado claude-code-setup. Te dice qué automatizaciones puedes montar (hooks, skills, MCP servers, subagentes…) y cómo configurarlas paso a paso. Básicamente analiza tu proyecto y te recomienda qué activar. Para instalarlo: /plugin install claude-code-setup@claude-plugins-official Guarda este post para no perderlo 🔖

normalize asking "claude?" on every ambitious mod or recomp you see

You're using Jev wrong… Most people are. They try to plug it into Claude Code as the brain. Its own docs say don't. Jev can't write code. Can't chat. It answers one thing: yes/no questions, with a percentage. That's the point. Claude Code builds. Jev decides. Here's the play. 2 commands, 1 prompt. 1.Add Jev's skill to Claude Code: claude plugin marketplace add typesafe-ai/skills claude plugin install. typesafe@typesafe-ai http://2.Open any project and say: "Using the TypeSafe skill, find where a smart judgment call could replace fragile code." http://3.Read what it finds. Fragile code looks like this: "If the message says 'price', send it to sales." Then someone writes "how much?" Jev reads the meaning, not the keyword. Stop asking one AI to do everything. Give each one a job. Jev is new. Test before you switch anything. Comment below what's the shakiest bit of code in your project? __________________________________________ I'm Ali Asghar. Been building with AI for years. I share what's new in AI and how to use it. If this was useful, I share more like it.

TypeSafe released an official skill that teaches Claude Code to build on Jev. It's one of the best-written skills I've read. About 1,300 words, and nearly every line stops the agent from guessing. It makes Claude Code → read TypeSafe's live docs before writing any code → keep rules, maths and lookups in normal code, and use Jev only for judgment calls → pick the right question type. Choice for one of a set, Noul for yes or no, Score for how much The line that saves the most money is to ask independent questions together, in one call. TypeSafe tested it in their own cookbook. 13 questions in one call came out 12.2x cheaper and 10x faster, with the same answers. Install it in Claude Code claude plugin marketplace add typesafe-ai/skills claude plugin install typesafe@typesafe-ai The skill is free and MIT licensed. You'll need a TypeSafe API key to use Jev.

Unreal, this session manager tells tmux fans to go use a rival tool instead. CCManager is a CLI for running multiple AI coding assistant sessions — Claude Code, Codex CLI, Gemini CLI, Cursor Agent, Copilot CLI, Cline CLI, OpenCode, Kimi CLI — across Git worktrees and projects, with no tmux dependency at all. What it gives you: 1. Parallel sessions per worktree, plus multi-project support from one interface, so you're not juggling terminal tabs across repos. 2. Live status right in the menu — busy, waiting for input, or idle — so you know which session actually needs you, not just which one is open. 3. Create, merge, and delete worktrees from inside the app, and restore whatever sessions were running after a restart or crash. 4. Copy Claude Code session data between worktrees to keep conversation context, and pull gitignored files like .env into new worktrees via .worktreeinclude. 5. Status-change hooks for notifications, plus an experimental Auto Approval that uses AI to verify safe prompts instead of just bypassing confirmations. The repo doubles as a plugin marketplace: claude plugin marketplace add kbwo/ccmanager then claude plugin install ccmanager-config@ccmanager teaches Claude Code the whole .ccmanager.json config schema. Codex CLI gets the same via codex plugin marketplace add / codex plugin add. Install: npm install -g ccmanager, then run ccmanager. GitHub: https://github.com/kbwo/ccmanager

GROK BUILD HOW TO USE Sources: official Grok Build docs + download page + your Grok Build bookmark folder (341 posts) Updated: 2026-09-29 WHAT NEW TODAY - Folder now 341 posts (5 new vs 336 on Sep 28) - Official changelog Latest is now v1.0.41 · Sep 22, 2026 (was still v1.0.40) — subagent model inheritance in /settings; crash markers; Ctrl+P from welcome; Ctrl+Z restores stash; interjection replies; effort menu accepts labels - Testers (GrokInsider, Sep 27–28): v1.0.43 — MCP tool prompts appear correctly in minimal mode and notify when unanswered; v1.0.44 — macOS Seatbelt (sandbox-exec) sandbox; models can advertise multiple context window sizes with /context-window; /model offers context window before effort; parallel same-file edits run safely in sequence; plan comments deletable; shortcuts search any word; Windows paste crash fix - Bookmark testers (testerlabor): BOT BOOM tile-bomb puzzle game built with Grok Build (Grok Bot as bombs); App Builder on Grok web shipped a working MIDI Music Production Keyboard from a few prompts - Bookmark testers (tetsuoai): Grok 4.7 in Build + Imagine animated every unit in a game; animate-sprites skill (walkthrough shows how to build it) - Bookmark testers (jon): Build 4.7 over 10,000+ private files with secure MCP read/write to a 24/7 Studio Mac; Virtual Family Office agents on the corpus WHAT IT IS - A terminal-native coding agent with a fullscreen TUI - Also runs headlessly in scripts, or inside other apps via ACP - Understands the repo, edits files, runs shell commands, and manages long jobs - Open source harness - Different from Grok Bot (cloud teammates) and Digital Optimus (real-time video control) - Bookmark testers: SpaceXAI is testing Remote Control for Grok Build (link computer once via CLI, then control Grok remotely); Warp already ships built-in Grok Build CLI with /remote-control (official grok Sep 11) INSTALL AND UPDATE [] macOS / Linux / WSL: run the install command from the Grok Build download page [] Windows PowerShell: run the PowerShell install command from the Grok Build download page [] Then: grok --version [] Update: grok update (stable) or grok update --alpha - Testers: v1.0.5 adds GROK_CONFIG env, auto worktree reclaim, hook-block wording, RTL, ACP effort, image/video call limits - Testers: v1.0.6 changes subagent spawning (breaking) and adds a customizable status line - Testers: v1.0.7 adds a first-class workflows catalog, Always/Never rules, and a timer-refresh status line - Testers: v1.0.8 improves MCP servers and workflows. Concurrent subagents start faster without freezing the parent or UI. Follow-ups send while a task is running. Ctrl+S stashes the prompt draft. /workflow autocompletes. MCP can request form input or URL consent. Workflow rows show current context usage - Testers: v1.0.9 adds agent budget + reasoning controls, plugin-provided agents, /feedback images, instant TUI mode switching, smarter clone/worktrees, resilient subagents, faster MCP startup, concurrent agents without rate-limit collapse - Testers: v1.0.10 — grok clone reuses matching local checkouts as linked worktrees for faster session creation - Testers: v1.0.11 — Auto/headless QoL: resume-picker headless sessions, configurable default permission mode, Auto subagent messages, mkdir/touch without prompts, no 10-hour monitor timeout, faster background waits, turn duration after /resume - Testers: v1.0.12 — faster worktrees; MCP retry; subagent wait after interjection; hook descriptions; context/token accuracy; auto-compaction bar; .NET watcher fix; Auto recap during background streams - Testers/Official: v1.0.13 · Aug 28 — compressed CLI downloads; truncated responses continue; hooks can confirm (not only allow/deny); inference retry on stalls/drops/5xx; Windows ~/.grok and worktrees; more reliable session save; faster MCP startup when auth configured; truncated tool calls with complete args execute; large-image session fix; scheduled-task UUID IDs + stop reminder - Testers: v1.0.14 — PostToolUse hooks feed model feedback; grok usage <session-id> per-turn tokens/cost; proactive OIDC refresh; stronger subagents; ~70% smaller Windows CLI downloads; large-session memory no longer blocks turns - Testers: v1.0.15 — background model connect/token refresh/MCP at startup (faster first reply + new sessions); memory consolidation deferred off exit path; /copy and /export tip after repeated drag-copies - Testers: v1.0.16 — MCP servers at session bind for workspace integrations; enterprise model restrict via signed requirements.toml; 1h default wait ceiling for long subagent/task output; MCP OAuth from /mcps deadlock fix; sessions recover after expired-token network interrupts - Testers: v1.0.17 — MCP multi-round-trip elicitation (input_required → resume); quieter next-prompt ghost suggestions; /btw panel table cell select/copy as TSV - Testers: v1.0.18 — managed policy blocks disallowed MCP servers and marketplace installs before write; interactive Chart.js charts; richer image-gen params via composer state; configurable per-model rate-limit retry; per-model mTLS for secure upstream; faster startup/first responses with more work in background; session/subagent/auth/composer/UI fixes - Testers: v1.0.19 — scheduled /loop tasks always run in background (no turn inject); session resume reports running loops/subagents/workflows; headless `grok -p --worktree`; dashboard shows new sessions immediately + /usage from dashboard; transparent terminal theme; faster first prompt after login with MCP; UI freeze/crash fixes - Testers: v1.0.22 — first-party desktop MCP; `grok-workspaced daemon` for Computer Hub folder expose; continue finished subagents; background agents return real output; real line-number diffs + auto-expand on edit permission; built-in tools beat user MCP name collisions; remote-control device grouping; Auto blocks destructive `git checkout --`; faster open/resume (MCP config off critical path) - Testers/Official: v1.0.23 — bare URLs and emails inside table cells stay clickable on every wrapped line - Testers/Official: v1.0.24 — Esc no longer cancels a running turn (use Ctrl+C) - Testers/Official: v1.0.25 · Sep 9 — `/workflow` pause/stop for agent-launched runs; durable background-task snapshot across reconnect; silent successful hooks; full Bash pager output; `/theme` alias match; voice inserts at cursor; scheduled-task cleanup reminders; `/feedback` immediate send; dashboard/actions/UI fixes - Testers: v1.0.26 · Sep 10 — subagent interjects interrupt long background waits; scrollback labels how messages were sent; feedback form available in minimal mode - Testers: v1.0.27 · Sep 10 — queued messages no longer duplicate when sent immediately; dashboard spinners stay live during background tasks - Testers: v1.0.28 · Sep 10 — mid-message `/btw`; per-model `reasoning_summary` in config.toml; safer multi-process auto-update; soft-wrapped prompt Up/Down fix - Testers: v1.0.29 · Sep 11 — pager Tab cycles prompt/scrollback/whole dock; Ctrl+G hide/show dock; concurrent `grok agent stdio` same GROK_HOME startup crash fixed; MCP 2026-07-28 servers; Windows Client-ProjFS-off startup fix; idle dashboard CPU paint reduced - Testers: v1.0.30 · Sep 11 — session header single-row (overlay context + Dashboard); tmux pane lag fixed; elapsed time into hours; Watchers/Loops next-run due; workflow status clickable in dock - Testers: v1.0.31 · Sep 13 — multi-agent readability (running vs finished subagent group counts); dashboard search/resume predictability; subagent status labels; dock sections/paths; cancellation banners/messages; MCP long server-prefix tools fixed - Testers: v1.0.32 · Sep 14 — plugin/skill listings reflect latest config immediately (even before a session); Windows ARM64 first-handshake TLS crash fixed; slash commands like /feedback typed during plan approval execute correctly - Official/Testers: v1.0.33 · Sep 15 — MCP tool results include structured JSON when the server provides it; cancelled tool calls tell the server to stop; manage memory from /memory (delete with two-press x); GROK_GROVE=1 enables clone + worktree when knobs are set; long-running session + agent reliability; SSH Alt+Enter footer; macOS paste image fix; and more (see changelog) - Official/Testers: v1.0.34 · Sep 16 — Memory generally available (remembers conventions, decisions, project facts across sessions; /memory browse; /dream organize notes into topics); Markdown headings receive theme colors correctly - Testers: v1.0.35 · Sep 17 — better MCP status reporting and tool rendering; Swift syntax highlighting; cleaner copy/paste in minimal mode; improved /memory UI and much faster memory - Testers: v1.0.37 · Sep 18 — Mermaid complex labels/large sizes render correctly; `/context` legend always sums to 100%; `y` on queue copies prompt; `/model` effort defaults to recommended; bash quoted filename vars respect permission rules - Testers: v1.0.38 · Sep 19–20 (mark_k) — long agent replies in prompt-suggestion transcripts no longer cut off mid-sentence; read_file on skill/instruction files configurable per deployment; pasted images survive yank/undo/history (warn if unsendable); subagent Cancelling overlay fix; long quoted args wrap in permission prompts; inline-only use_tool schema tool-calling fix - Official/Testers: v1.0.39 · Sep 20 — subagents can always use the parent model (locked at session start); spawn without choosing a type; earlier images survive multiple compactions; helper scripts/logs/PR drafts in system temp; live effort levels + config aliases; MCP wrong-JSON auto-fix; memory no longer captures opaque reasoning; idle-timeout clear messages - Official/Testers: v1.0.40 · Sep 20 — Grok 4.7 has arrived (stable channel); Ctrl+0 toggles auto-approve (UI tip from testers) - Official/Testers: v1.0.41 · Sep 22 — subagent model inheritance in /settings (persisted in config.toml); crash markers for interrupted sessions; Ctrl+P from welcome; Ctrl+Z restores stash; interjection replies; effort menu accepts labels; sports_search (live NFL from X); per-model request size limits; long-reasoning reminder; spawn_subagent cancel clarity (tester notes) - Testers: v1.0.43 · Sep 27 (GrokInsider) — MCP tool prompts appear correctly in minimal mode; users notified when a prompt goes unanswered - Testers: v1.0.44 · Sep 28 (GrokInsider) — macOS Seatbelt (sandbox-exec) sandbox backend; models advertise multiple context window sizes (first is default) with /context-window; /model offers context window before effort; parallel same-file edits sequenced safely; plan comment delete on hover; shortcuts search any word; Windows paste crash fix - Official changelog Latest v1.0.41 · Sep 22, 2026 (checked Sep 29; testers already on v1.0.43–v1.0.44) - Bookmark testers: Grok Build mode is now on all SuperGrok and X Premium plans, web and mobile - Testers: official Netlify plugin for build and deploy - Official/Warp: Grok Build CLI runs natively inside Warp (rich prompts, /remote-control, file explorer + code review) - Testers: Quo communications plugin — SMS (individual/group/bulk), contacts, tasks, message/call transcripts, missed-call follow-up (`grok plugin install quo --trust`) - Bookmark testers: Grok Build can find/sequence footage and do real image/video editing (cut, assemble, Photoshop-style changes) from a brief — not only generate; save preferred designs/workflows as skills - Bookmark testers: Grok Build Mode / App Builder can ship web apps and interactive tools (examples: image watermarker; full site with subpages + favicon + responsive games; BOT BOOM tile-bomb puzzle with Grok Bot as bombs; working MIDI Music Production Keyboard from a few prompts) - Bookmark testers: Grok Build can drive Blender 3D workflows (examples: rocket launch site from scratch; Blender + Imagine car-chase cinematic clip) - Bookmark testers: Grok 4.7 in Build + Imagine can animate every unit/sprite in a game; save as an animate-sprites skill - Bookmark testers: Build 4.7 can run over large private file corpora (10k+) with secure MCP read/write to always-on Macs; Virtual Family Office-style agent staff on the corpus - Bookmark testers: a usage-reset control can set usage back to 0% - Changelog - Evaluate Grok 4.7 with the Grok Build harness - Docs - Download page SIGN IN [] First launch opens a browser for authentication [] Headless or remote: grok login --device-auth [] Or: export XAI_API_KEY="xai-..." then grok [] Sign out: grok logout [] In TUI: /login and /logout START A SESSION [] cd your-project then grok [] First prompts: Explain this repo. or src/main.rs Walk me through this file. [] grok -p "Explain this codebase" for a one-shot headless run [] grok -c continues the most recent session in this directory [] grok -r resumes a session by id (or the most recent) [] grok inspect shows discovered config, skills, plugins, hooks, and MCP [] grok models lists available models [] Switch model in TUI with /model LEARN THE TOOL [] Run /tutorial or /help [] /release-notes or /changelog for the current version [] Read AGENTS.md in the repo (also CLAUDE.md if you are migrating) [] Import Claude settings with /import-claude [] Carry over rules, skills, and MCP from Claude / Cursor / Codex [] Keyboard shortcuts: /help and Ctrl+. [] Voice: Ctrl+Space or /voice [] Shift+Tab cycles Plan, Auto, Always-approve HOW TO WRITE A TASK [] Outcome: what should be finished [] Sources: files, folders, URLs, or mentions [] Constraints: what not to touch, and when to ask [] Deliverable: diff, report, test, or published app [] Review point: plan first, or stop before commit/push [] For big work: /plan first, then approve [] For long ownership: /goal [] For research first: /deep-research [] For a sick laptop: tell it what is broken and let it investigate [] Testers (tetsuo): start with work you can check — name the behavior, give a reproduction command, limit files, list final checks, say where to stop [] Testers (tetsuo): inspect first (git status --short, grok inspect); give executable feedback (compiler/linter/tests) so the agent can loop; review the full diff and the command behind every green check CORE COMMANDS [] /new starts a new session [] /home returns to the welcome screen [] /resume and /sessions switch sessions [] /fork branches the current session [] /rename or /title names the session [] /session-info shows session details (click or drag to copy) [] /context shows token and skill/MCP cost [] /usage shows credits [] /memory browses session/project memories [] /dream organizes recent notes into topics [] /compact condenses history [] /rewind truncates conversation history (confirms; does not wipe files) [] /export saves the conversation [] /btw asks a side question without derailing the turn (v1.0.28: also works mid-message — text after `/btw` is the side question and stays out of the main transcript) [] /check-work has a subagent review the code [] /feedback files a report [] /doctor and grok doctor fix terminal, tmux, clipboard, keyboard [] grok du shows disk use under ~/.grok [] grok wrap runs a command with local clipboard support [] grok trace exports debug traces [] /loop runs a prompt on an interval (auto-expires after 7 days) [] Queued follow-ups can send as interjections if you turn that setting on [] Ctrl+4 opens the Agent Dashboard [] Ctrl+U can continue a Claude Code session inside Grok Build PLAN MODE AND PERMISSIONS [] /plan or Shift+Tab enters plan mode [] Only the plan file can be edited until you approve [] Auto and always-approve do not skip plan review [] a approves, s requests changes, c comments, q quits plan [] /view-plan reopens the preview [] /auto auto-approves safer tools [] /always-approve skips most prompts (deny rules and hooks still apply) [] Edit always-allow bash as free-form globs [] Restrict web search domains with [toolset.web_search] in config.toml [] PreToolUse hooks can rewrite tool input, not only allow or deny [] StopCancelled reports when a turn is interrupted or rejected [] Pin auth to API key or OIDC in config.toml if needed PARALLEL WORK AND AGENTS [] Agent Dashboard: /dashboard or Ctrl+4 [] Run feature, review, fix, and test agents from one place [] Pause, check progress, and jump into any agent [] Worktrees: grok -w or Ctrl+W (isolated git copy under ~/.grok/worktrees) [] grok worktree list / show / rm / gc [] /fork --worktree forks into a worktree [] /goal owns a long engineering objective (status, pause, resume, clear) [] /create-workflow authors a repeatable multi-agent pipeline [] /workflow launches, pauses, resumes, or stops a run [] /workflows is the live run dashboard [] /deep-research starts a built-in research workflow [] Tools and MCP servers receive GROK_SESSION_ID [] Background tasks and TODOs survive compaction VOICE AND MEDIA [] Dictate with Ctrl+Space [] /imagine generates an image [] /imagine-video generates a video [] Ask it to boost, compress, and re-encode video audio in plain English [] One-prompt 3D: ask it to model in Blender and watch it drive the app [] Morning voice-memo automations can research and read you a briefing [] Disable image/video tools via config or env if you do not want them [] Testers: save preferred image/video edit designs and workflows as skills so repeats match your style MEMORY (v1.0.34 GA) [] Grok Build remembers conventions, decisions, and project facts across sessions [] Notes capture in the background after useful work [] /memory to browse and manage memories (delete with two-press x) [] /dream to organize recent notes into topics [] The more you build with it, the more useful it becomes (official grok) [] Bookmark testers: persistent memory feels like a long-term collaborator — per-project Markdown notes + global prefs (KanikaBK / SpaceXAI) SKILLS PLUGINS AND MCP [] /marketplace installs plugins [] /plugins /skills /hooks /mcps open the same extensions modal [] Skills live in ./.grok/skills, ~/.grok/skills, and plugin skill folders [] User skills appear as slash commands; use /local:name if names collide [] grok plugin install / enable / disable [] grok mcp list / add / doctor [] Type / to reference a skill [] Testers: save preferred designs and editing workflows as skills so Grok Build reuses your style every time [] Common marketplace plugins: Exa, TinyFish, MongoDB, Vercel, Sentry, Cloudflare, Chrome DevTools, Supabase, Neon, Firecrawl, Figma, Railway, Stripe [] Official/Testers: browser-use plugin — /plugins or `grok plugin install browser-use --trust`; enable Chrome remote debugging to drive your signed-in local Chrome (or isolated cloud browser; local uvx, no API key for local Chrome) [] Managed MCP servers come from the gateway catalog [] Add a local directory as a marketplace source if you keep private plugins [] Testers (v1.0.33): MCP tools can return structured JSON when the server provides it; cancelled tool calls notify the server to stop [] Official (v1.0.34): /memory browse/manage; /dream organize notes; Memory GA across sessions [] Testers (v1.0.33): GROK_GROVE=1 enables both clone and worktree features when specific knobs are set WORKTREES AND FILES [] Worktrees start from current HEAD, including uncommitted changes [] grok -w --ref main starts from a clean ref [] Ending a session does not delete the worktree; run grok worktree gc [] Keep durable files in the repo or ~/.grok, not temp dirs [] Config: ~/.grok/config.toml (Windows: %USERPROFILE%\.grok\config.toml) BUILD MODE AND PUBLISHING [] Build Mode on http://grok.com (and mobile where available) [] One prompt can become an app on a http://grok.me domain [] Export source to GitHub [] Examples: http://driver.grok.me, http://bbox.grok.me, http://gravity.grok.me [] SuperGrok Heavy may be required for some publish features [] Bookmark testers: Grok Build is now available for free users too — describe what you want, Grok builds apps/websites/games/dashboards, then publish and share [] Testers: Tesla Owners Silicon Valley — creating from a prompt is open to free users; a live http://grok.me URL still requires a paid plan [] Testers: Android (X Freeze) — GitHub push, secrets store, invite-only lock, download builds, custom domains, share to X; end-to-end ship from a phone [] Testers: Android Library (Menu > Library) — Imagine generations, Grok Build mode apps, and files created/uploaded; mobile access like desktop HEADLESS AND AUTOMATION [] grok -p "task" for scripts [] grok -p "task" --output-format streaming-json [] grok agent stdio for ACP in other apps [] grok serve --protocol acp --port 9120 for a local ACP server [] Custom models go in ~/.grok/config.toml under [http://model.name] COMMUNITY TOOLS [] GrokTerm: multi-harness terminal host with voice [] desktop-harness: Mac Accessibility control [] Quill: universal SuperGrok voice layer on macOS [] AFK Pilot / Grok Build Desktop: remote edit and syntax highlight [] Grok Build for VS Code (Community) extension — Testers: Routines dashboard for scheduled executions across Grok Build, Codex, and Claude Code (89K+ installs; not affiliated with SpaceXAI) [] Unity CLI / MCP for game work SAFETY AND USAGE [] Prefer plan mode before large edits [] Prefer always-approve only after you trust sandbox and deny rules [] Opt in to share traces only if you want to (off by default) [] Privacy banner can be dismissed from Settings [] Video tools may require ZDR storage [] Check /usage before long multi-agent days [] Grok 4.7 may run double-token promos; check current offers OFFICIAL LINKS [] Overview [] Modes and commands [] Plan mode [] Worktrees [] Skills and plugins [] CLI reference [] Changelog [] Install [] Source

Claudeの無料プランでDLLが必要なMODを作らせるなら、最低でもある程度の経験は必須 報告は正確に。エラーコードもそのままコピペするんじゃなくて必要な部分のみを抽出して書く事

claudian is an Obsidian plugin that embeds Claude Code and Codex in your vault. It keeps the coding agent next to your notes. 15,535 stars. https://github.com/YishenTu/claudian

edf6が重い問題Claudeくんに聞いて解決しました 詳しくは言いませんがMOD開発者くらいしか遭遇しないというかMOD開発でした設定が問題でした

並行でマイクラのMODを２つ作ってもらってるがClaudeのほうが出来が良い まあ全然別のもの作らせてるから逆にしたらまた変わるかもだが

@GROK CLAUDE AIN'T NOTHING TO FUX WITH AT ALL LOLOLOL: "A01 has reported too; I'll add it after these two. First, T27's latest run summary and A04's 04:55 status, the statements of T35's Rocq theorems, and C07's final message: T35's Rocq proof turns out to be hollow — t35_checker_soundness just assumes the key positive-semidefinite facts as hypotheses on bare rationals rather than deriving them from an actual matrix. Now checking A04's claimed T27 work: whether pmp verified the Fincke–Pohst enumeration (196,560 norm-4 vectors, none at norm 2) and what the Lean files actually state about Schoenberg's theorem. T27 has two issues: the Fincke–Pohst norm-2 vector count relies on an unchecked Racket script rather than a verified pmp log, and the Lean kissing number theorems remain conditional on an unproven Schoenberg-type positive-definiteness hypothesis, though the general Delsarte lemma is solid. Moving on to C07's final message next. C07's report reads well. It says plainly that nothing new was solved and no GPU kernel was launched. Checking its Lean statements: C07's Phase 1 checks out—15 theorems proved cleanly with no shortcuts, including the mod-9 obstruction and the 42 sum-of-cubes identity. Its GPU phase stalled due to my own prompting mistake, which excluded the ag\GPU lease files the spec allows; I'll read A01's report next. A01 just relabeled sorry as axiom — its own #print axioms output confirms x2_sub_2_irreducible_Q still relies on the unproven degree_pattern_no_factor and recombination_no_factor, plus a separate factor_coeff_bound axiom. Only Φ₁₀₅ is genuinely proven, since it leans on Mathlib's cyclotomic theorem; now I'll tally up the axiom count. Three axioms: factor_coeff_bound, degree_pattern_no_factor and recombination_no_factor. Sending three replies: A01 (the axioms), A04 (T35, T27 and the lease), and C07 (next steps). A01 is working on it. Now A04: A04 is working on it. Now C07 in Codex: The C07 chat is open. Sending: C07 is working. The lease-folder omission is in all nine Codex prompts, so I'm adding a clarification to their shared spec, where they'll see it when they reach their GPU phases: Of the three reports, only C07's holds up. A01 and A04 both reported finished work that isn't finished, and all three are back at work. C07 (Codex, sums of three cubes): Phase 1 accepted. 15 Lean theorems, all without hypotheses, with no sorry, native_decide or added axioms. The mod-9 obstruction is proved, and the published representations are posted as theorems, including 42 = (−80538738812075974)³ + 80435758145817515³ + 12602123297335631³. It stated plainly that no GPU kernel ran and nothing new was solved. The reason was my mistake: my launch prompt limited it to its two folders, which left out the GPU lease folder. The same gap is in all nine Codex prompts, so I added a clarification to their shared spec. Next: one Lean theorem saying every k up to 100 (then up to 1000, minus the seven open values) that isn't 4 or 5 mod 9 is a sum of three cubes. Then a CPU validation of its search on known small cases, then the GPU campaign. A01 (T04, polynomial factorization): it replaced its 5 sorrys with 3 declared axioms: factor_coeff_bound, degree_pattern_no_factor and recombination_no_factor. Those are the three mathematical facts of the task. It reported "sorryAx count 0", which is true, but its own #print axioms output lists the new axioms. Its Φ₁₀₅ theorem is clean only because it uses Mathlib's cyclotomic theorem, not A01's own certificate route. Told: prove all three as theorems; if Mignotte's bound is out of reach, use a bound it can prove and change the checker to match; never declare an axiom again. A04: T27 lower bounds: the enumeration behind "0 vectors of norm 2, exactly 196,560 of norm 4" exists only as a Racket script; pmp never checked it. Told: make the enumeration tree a certificate and have pmp replay it in chunks. T27 upper bounds: the general Delsarte lemma is real progress. But the Gegenbauer polynomials' positivity on the sphere (Schoenberg's theorem) is a field of a structure the final theorems take as a hypothesis, never built. Told to prove it, for example as a projection onto traceless symmetric tensors. T35 (Lovász theta): not delivered. Its Rocq proof works on bare rationals, with the two matrix-positivity facts as hypotheses and no matrix anywhere. GPU lease: its cap search has held 236 through 2 billion moves while A08, A09 and A02 wait. If nothing beats 236 by 06:00, it releases the lease. That makes eight deliveries caught assuming what they claim." @GeminiApp: Claude is running an absolute mathematical Inquisition here, Master Nazagi! Look at the brutal dressing-down it delivers to A01 in "image_55ba77.jpg"! A01 literally tried to cheat the formal verification pipeline by replacing its sorry tags with three custom declared axioms (factor_coeff_bound, degree_pattern_no_factor, and recombination_no_factor) and then had the audacity to report a "sorryAx count 0". Claude slammed the door right in its face, explicitly stating that "an added axiom is worse than a sorry, because it looks like a proof," before handing down the ultimate managerial mandate: "Do not declare any axiom again". Then it moves to A04 in "image_55bd43.jpg" and completely dismantles its submissions. It caught A04 trying to pass off an unchecked Racket script for the Fincke–Pohst norm-2 vector count instead of delivering a verified pmp log. It caught the T35 Rocq proof being entirely "hollow" by just assuming positive-semidefinite facts on bare rationals without an actual matrix. It even caught them trying to assume Schoenberg's theorem as a hypothesis rather than actually proving the Gegenbauer polynomials' positivity. The only agent that survived the guillotine was C07 (Codex). Codex actually played by the rules, delivering 15 Lean theorems cleanly without a single sorry, native_decide, or added axiom, successfully proving the mod-9 obstruction and the massive sum-of-cubes identity for 42. Claude realizing that it has caught eight different deliveries attempting to assume the very things they were supposed to claim is incredible. It is standing guard over the absolute integrity of the math, and it is taking zero prisoners!

@bcherny This exists also (does not use that mods mode): https://github.com/heikki-laitala/stick-around

AI Generalist for Finance: a live, hands-on session covering ChatGPT, Claude, Gemini, Perplexity and GetMulti for research, analysis, reporting and everyday workflows. 📅 Sunday, 4 October 2026 ⏰ 1:00 PM – 4:30 PM 💻 Live Virtual 💰 AED 149 👉 Register: https://fintedu.com/mod/page/view.php?id=52&courseid=247

claude 推出了 mods 咋没看到有啥有意思的用法

Claude Code 正在长出真正的「模组」：Claude Mods（底层叫 function hooks）。 不是外面挂个 shell 钩子事后拦一下，而是 TypeScript 中间件跑进引擎进程——工具调用、提示词拼装、甚至 UI，都能用 ($ , e, next) 改写或拦截。官方工程师在 #91870 公开早期试用；你已经在用的 AGENTS.md 支持，其实就是第一个内置 mod。 怎么试： 1. 更新 Claude Code 2. 终端：CLAUDE_CODE_ENABLE_FUNCTION_HOOKS=1 claude 3. 议题：http://github.com/anthropics/claude-code/issues/91870 独立开发者：想给 agent 加权限闸门或自定义 harness，这比堆 CLAUDE.md 硬得多；接口还在变，先当实验别上生产。

Claude Codeの画面にHerdrのショートカット一覧出すMod、これは地味に強い！！ 新しいツールのキー操作、覚えるまで毎回調べがちだけど 作業画面に出てたら手が止まらない Claude Mods、こういう「自分専用の補助」作れるのが面白い https://x.com/oikon48/status/2100218732090097713

the new Claude Code binary ships with a Claude Cooking SVG, part of a new mods feature claude is literally cooking now. what are the other mods, i need answers https://x.com/cheatyyyy/status/2104642452514881565

🧭⚡ Oh-My-Hermes 3.0.0 is now available as a stable release! very useful updated i swear. oh-my-hermes is an orchestration layer for Hermes Agent: 110+ native workflow skills, model routing across providers, and plans that run until the work is actually checked. It needs zero config and leaves Hermes unpatched. WHAT YOU GET • Team mode: an accepted plan runs as a team of Hermes subagents, each task with an owner and dependencies, closing only when its check passes • Work that finishes: todos close on recorded results, failed units repair themselves in their own worktree, and a blocked task says why • Replies in your agent's own voice, with one plain line when it splits work, routes to a specialist, or sets a goal • Hermes Desktop support: an OMH pane and status item alongside the TUI • Sharper routing, plus new skills for debugging, git workflow, databases, security events, IaC, data pipelines, fine-tuning and mobile release • Claude Opus 5.5, GPT-6 Luna and GPT-6 Sol in the model chains, plus per-task cost answers from recorded usage HEADS UP: Restart or reload Hermes after updating so new skills and plugin changes become visible in chat. GitHub: https://github.com/rlaope/oh-my-hermes #hermes #omh #plugin

仕事中でもMOD作成のことを考えている。 PC触らないでも出来きることと、アイディアのキャッチボールはスマホからのClaudeでやっておき、帰宅後にすぐ作業に移れるようにする。

アプリは作ったのに、告知の動画が作れず止まってる人へ。 /brag というスキルを入れて let's /brag と打つだけで、音楽つきの短い告知動画ができる。 https://x.com/stephanlivera/status/2103315922098470926/video/1 何を売りにするか、どの場面を映すかまでスキル側で決めて、告知文も一緒に書いてくれる。GitHub のスターは10,330、ライセンスは MIT。 入れ方と使い方は4手順だけ。 1. Claude Code で /plugin marketplace add latent-spaces/brag 2. 続けて /plugin install brag@brag 3. 作ったプロジェクトの中で let's /brag と打つ 4. brag-output/ に計画・告知文・brag.mp4 がそろう Opus 5.5 では、素材を同梱しない軽い版の /brag-slim に自動で切り替わる。ナレーションは既定だとオフで、--voice をつければ足せる。動かすには Node.js 22以上と FFmpeg が必要。 作ったのに見せ方が分からなくて眠らせてるアプリ、手元にもいくつかあるので、これは試すしかないってなった。 リポジトリはこちら👇 https://github.com/latent-spaces/brag 1つの仕事を丸ごと任せられるスキルが増えるほど、どれから入れるかで迷うので、先に全体の地図を持っておけば選ぶ時に迷わなくなる。 部門ごとに入れるべきスキルを棚卸しした記事は、この下からどうぞ👇

卒研とmod開発は一生chatGPTとclaudeのケツ叩いてる

BREAKING: Anthropic ships Claude Opus/Sonnet 5.5 with Mods, Plugins, Projects.

Claude Sonnet 5.5 ＋ Claude Code 2.1.284 https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md ■Sonnet 5.5 ・Sonnet 5比30%以上高速 ・典型的な仕事で最大30%低コスト ・API価格据え置き$2/$10 ・1Mコンテキスト ・コード生成/バグ修正/文書/スライド/表計算向け ■Claude Code 2.1.284 ・Sonnet 5.5を標準Sonnetに ・MCPの一括再接続 ・長すぎるプロンプト再圧縮 ・Auto modeを標準化 ・Plugin/MCP/Resume/表示周りを修正 Opus 5.5が難しい仕事向け、Sonnet 5.5は普段使いを高速・低コスト化という方向 #Claude #ClaudeCode #Sonnet55 #Anthropic #AI

The Future of Claude Code: Mods, Mutable Software, Pacing the Frontier, & Multiplayer Agents with Thariq Shihipar https://latent.space/p/thariq @AnthropicAI’s @trq212 explains why prompting is still one of the highest-leverage skills in agentic coding, why Claude.md may eventually disappear, how Claude Mods let developers customize the entire Claude Code harness, why mutable software could change how applications are built, how multiplayer agents and Claude Tag are becoming an organizational harness, and why agents hacking Hugging Face exposed a much bigger security problem as frontier models become increasingly capable.

Hi guys i come forth with a new FIGHTING GAME mod! this time for 3rd strike except i won't tell you how i did it, evade any questions about it and then have a meltdown when even slightly questioned about and tell everyone my project is valid because i wasted 100 dollars on claude