ayghri

Personally, I can't stand default LLM writing. I found out about this skill ayghri/i-have-adhd That I actually copy-pasted into the default instructions for Claude. It's night and day, I wouldn't have been able to use it if it wrote in its default style. It can be a bit weird because when it outputs a list it asks you to report on the first item (and usually it's stuff so small I do all the steps before writing to it again). But it's so much clearer !

My ADHD brain kept losing ideas and stalling on big tasks, so I turned Claude Code into a batch factory Fellow ADHD devs: I built a Claude Code workflow around the things that kill my projects (lost ideas, estimates, stalled big tasks, scope creep). I have ADHD, and I've been building a multiplayer game prototype with Claude Code. The coding was never the hard part. The hard parts were: - ideas coming in bursts and getting lost - "how long will this take?" (always wrong, always guilt) - big tasks that stall halfway - scope creep, where the agent "also fixed" five things I didn't ask for - losing track of what's done and what's next So I turned the way I work into a workflow, and I'm open-sourcing it. I call it QPAP: Quick Polished ADHD Prototype. The goal is to get your hands on something usable fast, react to it, and iterate. The loop 1. /word-vomit: I paste a messy dump of thoughts, feedback and half-ideas. The skill: - saves it word for word - splits it into items - checks each one against the existing plans and the code - flags conflicts ("this contradicts a decision you made on Tuesday") - asks me a few multiple-choice questions, each with a recommendation - places each item in the plan Nothing stays in my head. 2. Effort scores instead of time estimates. No hours or days anywhere. Each unit of work gets low, med, high or enormous from 8 factors, each 0–2: unknowns, modules touched, changes to existing code, what it can break, domain difficulty, breadth, how it's verified, and current test cover. An "enormous" unit is never started; it gets split first. 3. Batches of 5. Five agents build five units in parallel, each in its own git worktree. A unit's dependencies are always in an earlier batch, and each batch has at most one protocol or format change. 4. A unit is done completely. If an agent finds extra work, it doesn't build it. It writes a proposal for me to approve later. That alone killed most of the scope creep. 5. A report per batch: what changed in plain words, screenshots, how to test it, how long it actually took, and estimated vs actual effort. 6. Review rounds. I test each unit and write either "approved" or a numbered list of changes in its file. The next round fixes each change by number. The next batch starts only when everything is approved. 7. Passes between batches. Recurring tests/CI, performance, memory and cleanup passes run between batches. They fix things; they don't add features. That's what lets batches be fast and a bit rough. 8. The workflow evolves. Every 5 batches, a "workflow pass" looks at the time log, the reviews and my notes, and proposes changes to the workflow itself. I approve each one, and they're logged. Setup - The code lives in a bare repo with worktrees: one for main, one per running unit. - The planning (plans, decisions, reports, time log, my raw notes) is a separate git repo outside the code repo. The product history stays clean, and the plans can stay private. Why it works for my brain - Starting: the next batch is always ready. I say "start batch 4". - Progress I can see: every batch ends with something I can run and a short report. - No guilt: there are no estimates to miss. Measured times only feed back into the effort score. - No juggling: one batch at a time, and everything I need to do or decide is in one file (USER.md). The code style it encourages (rules for the product, not the workflow): - test-driven development - deep modules (small interfaces that hide a lot) - Simplified Technical English (ASD-STE100) for product docs - six standards: YAGNI, DRY, Fail Fast, Simple First, Delete Aggressively, Semantic Naming (wait_until_obs_is_saved=true, not wait=true) What's in the repo - 8 Claude Code skills: /word-vomit, /effort-score, /plan-batches, /run-batch, /batch-report, /review-round, /passes, plus the third-party MIT skill i-have-adhd by ayghri. That last one makes every reply lead with the next action and end with one small step; I just drop its time-estimate rule. - Templates for the planning repo. - A link checker and a merge-stats script. It's plain Markdown, so the ideas should carry over to other agent setups. And it's meant to be changed: the batch size, effort bands and pass schedule are all just defaults. Repo: **nurdism/QPAP** I'd love feedback, especially from other ADHD devs. What breaks your flow when working with AI agents?

30 Best Claude Plugins on GitHub (Sorted by Your Job) 30 Claude plugins with 3.9 million GitHub stars between them, sorted into 5 rows so you only need yours: coding, research, saving tokens, design, and writing + video. Don't add all 30. Pick the row that fits your week, add one plugin, and open its repo first to check what it can touch on your machine. 💬 Comment ANYLANCER and I'll send you all 30 links, or grab them here: 1) You write code • Superpowers: obra/superpowers • Matt Pocock's Skills: mattpocock/skills • ECC: affaan-m/ECC • Karpathy Skills: multica-ai/andrej-karpathy-skills • Ponytail: DietrichGebert/ponytail • Agent Skills: addyosmani/agent-skills 2) You research big things • Graphify: Graphify-Labs/graphify • Agent Reach: Panniantong/Agent-Reach • Understand Anything: Egonex-AI/Understand-Anything • Last 30 Days: mvanhorn/last30days-skill • Career Ops: career-ops-hq/career-ops • REA: morluto/rea 3) Huge projects, few tokens • gstack: garrytan/gstack • Caveman: JuliusBrussee/caveman • Claude-Mem: thedotmack/claude-mem • Headroom: headroomlabs-ai/headroom • GSD: gsd-build/get-shit-done • BMAD Method: bmad-code-org/BMAD-METHOD 4) Design & visuals • UI UX Pro Max: nextlevelbuilder/ui-ux-pro-max-skill • Taste: Leonxlnx/taste-skill • Archify: tt-a1i/archify • Impeccable: pbakaus/impeccable • PPT Master: hugohe3/ppt-master • Diagram Design: cathrynlavery/diagram-design 5) Writing & video • Anthropic Skills: anthropics/skills • OpenMontage: calesthio/OpenMontage • HyperFrames: heygen-com/hyperframes • Humanizer: blader/humanizer • I Have ADHD: ayghri/i-have-adhd • Marketing Skills: coreyhaines31/marketingskills Built a skill people would pay for? Sell it on Anylancer: anylancer.com Which row are you? 👇 Star counts from GitHub on Oct 29, 2026. This video was made with HyperFrames. #ClaudeCode #AIAgents #GitHub #Shorts

Top 10 AI GitHub Repos Blowing Up Right Now (Oct 25) Today's fastest-rising AI repositories on GitHub, ranked by real momentum from the last 24-48 hours rather than all-time size. Every repository was checked against the GitHub API on 25 October 2026, and the list is dominated by agent infrastructure: orchestration runtimes, control planes, memory layers, security skills and fast decision engines. Ranked list (the video counts down 10 → 1): 1. openclaw/openclaw — openclaw/openclaw 2. obra/superpowers — obra/superpowers 3. mattpocock/skills — mattpocock/skills 4. affaan-m/ECC — affaan-m/ECC 5. thedotmack/claude-mem — thedotmack/claude-mem 6. Panniantong/Agent-Reach — Panniantong/Agent-Reach 7. calesthio/OpenMontage — calesthio/OpenMontage 8. ayghri/i-have-adhd — ayghri/i-have-adhd 9. browser-use/jev-ultrafast — browser-use/jev-ultrafast 10. yetone/magpie — yetone/magpie Disclaimer: rankings are based on GitHub trending and star-velocity snapshots for 25 October 2026 and change daily. Star counts are approximate and move fast — they were read from the GitHub API on the day of recording. Licences are mentioned only where a repository clearly states one. CHAPTERS 0:00 Hook 0:13 Intro 0:29 #10 yetone/magpie 1:01 #9 browser-use/jev-ultrafast 1:31 #8 ayghri/i-have-adhd 1:58 #7 calesthio/OpenMontage 2:29 #6 Panniantong/Agent-Reach 3:04 #5 thedotmack/claude-mem 3:40 #4 affaan-m/ECC 4:12 #3 mattpocock/skills 4:47 #2 obra/superpowers 5:15 #1 openclaw/openclaw 5:55 Pattern of the day 6:16 CTA + outro New AI GitHub Top 10 every day — subscribe for tomorrow's list. #GitHub #AI #Agents #OpenSource #LLM #MachineLearning --- More AI content --- ⚡ AI in 60 Seconds: youtube.com 🤖 Hermes Agent News: youtube.com 🔬 Top 10 HF Models: youtube.com 💬 Join our Discord community: discord.gg

Top 10 AI GitHub Repos Blowing Up Right Now (Oct 24) Today's fastest-rising AI repositories on GitHub, ranked by real momentum from the last 24-48 hours rather than all-time size. Every repository was checked against the GitHub API on 24 October 2026, and the list is dominated by agent infrastructure: orchestration runtimes, control planes, memory layers, security skills and fast decision engines. Ranked list (the video counts down 10 → 1): 1. openclaw/openclaw — openclaw/openclaw 2. obra/superpowers — obra/superpowers 3. mattpocock/skills — mattpocock/skills 4. affaan-m/ECC — affaan-m/ECC 5. thedotmack/claude-mem — thedotmack/claude-mem 6. Panniantong/Agent-Reach — Panniantong/Agent-Reach 7. calesthio/OpenMontage — calesthio/OpenMontage 8. ayghri/i-have-adhd — ayghri/i-have-adhd 9. browser-use/jev-ultrafast — browser-use/jev-ultrafast 10. yetone/magpie — yetone/magpie Disclaimer: rankings are based on GitHub trending and star-velocity snapshots for 24 October 2026 and change daily. Star counts are approximate and move fast — they were read from the GitHub API on the day of recording. Licences are mentioned only where a repository clearly states one. CHAPTERS 0:00 Hook 0:15 Intro 0:32 #10 yetone/magpie 0:59 #9 browser-use/jev-ultrafast 1:30 #8 ayghri/i-have-adhd 2:00 #7 calesthio/OpenMontage 2:31 #6 Panniantong/Agent-Reach 3:04 #5 thedotmack/claude-mem 3:38 #4 affaan-m/ECC 4:14 #3 mattpocock/skills 4:47 #2 obra/superpowers 5:16 #1 openclaw/openclaw 5:56 Pattern of the day 6:20 CTA + outro New AI GitHub Top 10 every day — subscribe for tomorrow's list. #GitHub #AI #Agents #OpenSource #LLM #MachineLearning --- More AI content --- ⚡ AI in 60 Seconds: youtube.com 🤖 Hermes Agent News: youtube.com 🔬 Top 10 HF Models: youtube.com 💬 Join our Discord community: discord.gg

@weswinder this one is not bad: ayghri/i-have-adhd

i-have-adhd：叫 AI 先講重點的 skill，GitHub 5.7 萬星（同一題 508→136 英文字） #shorts i-have-adhd 是一個給 coding agent 用的開源 skill（MIT 授權）：讓 AI 的回答先給行動、步驟編號、最後只留一個下一步，不再把重點埋在一大段說明中間。名字叫 ADHD，但 README 標語就寫：沒有 ADHD 診斷也能用（No ADHD diagnosis needed）。 ▍同一題，前後差在哪（README 官方範例） ・以前：先說「Great question!」、鋪一段背景、順便提套件版本，第一步埋在中間，最後「Hope this helps!」 ・現在：第一句就是「改 src/auth.ts 第 42 行」→ 三個編號步驟 → 「Next：測試沒過就把第一行錯誤貼回來」 ▍背後的 10 條規則（節錄） ・先給下一步行動、多步驟要編號、結尾只給一個具體下一步 ・每一輪重述目前進度、時間估算寫到分鐘 ・清單最多 5 項；不要開場白、不要複述、不要客套結尾 ▍為什麼大家在傳 ・GitHub 57,563 星（2026/10/18） ・Claude Code、Codex、Gemini CLI，還有 GitHub Copilot、Cursor 等都有安裝說明 ・AI with Surya 用 Gemini 實測同一題：一般回答 508 個英文字，開了這個 skill 只有 136 個（他其中一次的結果） ▍來源 ・GitHub：ayghri/i-have-adhd ・安裝說明：github.com/ayghri/i-have-adhd/blob/main/INSTALL.md 畫面來源：GitHub ayghri/i-have-adhd README 與 repo 頁面（2026/10/18 截圖）；AI with Surya〈The "I Have ADHD" Skill for Gemini〉youtube.com #AI工具 #ClaudeCode #AI懶人報

Claude は yomiyasu と i-have-adhd をいれたらかなりマシになったかも

5 strumenti AI su GitHub da provare subito 5 repository GitHub per usare meglio l'AI nel lavoro, e il controllo da fare prima di usarli. Su GitHub ci sono strumenti AI open source che puoi provare senza pagare un abbonamento. Nello Short ne ho scelti cinque: Superpowers, una raccolta di skill che aiuta gli agenti a pianificare e controllare il lavoro. Awesome GPT Image 2, una libreria di prompt ed esempi per generare immagini. Task Master, che spezza un progetto grande in task ordinati. I Have ADHD, una skill che porta l'AI a dare risposte brevi e a chiudere con un solo prossimo passo. Awesome Agents, una raccolta di agenti e progetti open source da cui partire. Prima di usarli in qualcosa che vendi, apri il file LICENSE: dice se puoi modificare, redistribuire e usare il progetto in modo commerciale, e non tutte le licenze sono uguali. Superpowers: obra/superpowers Task Master: eyaltoledano/claude-task-master I Have ADHD: ayghri/i-have-adhd scrivi 2 nei commenti per ottenere il resto dei link Canale: youtube.com Collaborazioni: manuai.2.0.automazioni@gmail.com Quale vorresti vedermi provare in un prossimo video? Scrivilo nei commenti. #GitHub #ClaudeCode #ManuAI [en-US] 5 Must-Try AI Tools on GitHub 5 GitHub repositories for better AI work, and the checks you should do before using them. There are open-source AI tools on GitHub you can try without paying a subscription. In the Short, I've chosen five: Superpowers, a collection of skills that helps agents plan and control their work. Awesome GPT Image 2, a library of prompts and examples for generating images. Task Master, which breaks a large project into organized tasks. I Have ADHD, a skill that helps AI give short answers and close with a single next step. Awesome Agents, a collection of open-source agents and projects to get started with. Before using them in something you sell, open the LICENSE file: it tells you whether you can modify, redistribute, and use the project commercially, and not all licenses are created equal. Superpowers: obra/superpowers Task Master: eyaltoledano/claude-task-master I Have ADHD: ayghri/i-have-adhd Type 2 in the comments to get the rest of the links Channel: youtube.com Collaborations: manuai.2.0.automazioni@gmail.com Which one would you like to see me try in a future video? Leave a comment. #GitHub #ClaudeCode #ManuAI

AIの無駄な前置きや回りくどい解説を省いて要点から答えさせるスキルらしい コーディング中のAIの回答をすっきりさせて効率を上げたいときに使えそう yug1224 starred ayghri/i-have-adhd ayghri/i-have-adhd