# AI Practice Commons > A free, open collection for learning AI through useful work, for people who are not engineers. Learners build something small, check that it does the job, and make it part of their week. It holds Agent Skills (SKILL.md), concept cards, a resource catalogue and blueprints (small project cards), all linked into one graph. Women Defining AI looks after it. Code Apache-2.0, content CC BY 4.0. Early preview: nothing is released yet. Records are either `core` (reviewed by a person) or `garden` (open drafts, possibly AI-proposed, not yet reviewed). Garden items are marked below. Never present garden content as reviewed. ## Start - [Home](/): What the commons is, the four-step learning loop (imagine, make, check, repeat) and the skill for each step, and a ten-minute start. - [Start here](/start-here/): How to install the practice coach and Build Cycle skills, common install problems, and a first blueprint. - [Thinking with AI](/thinking-with-ai/): Why the skills ask the learner to go first, the research on productive friction, and a reading list. - [About](/about/): Who looks after the commons, licences, the garden and core layers, and the files AI agents can read. - [Partner with us](/partner/): How model labs, schools, nonprofits and employers can fund or use the commons, and what sponsors cannot buy. - [Build with us](/contribute/): How to contribute, from fixing a broken link to adding a concept card. ## Data - [graph.json](/graph.json): Every record as a node and every typed link as an edge (builds_on, evolved_from, related, parallels, practises, teaches, explains and more). - [catalogue.json](/catalogue.json): All catalogue resources with provider, type, audience, cost, login, licence and last-checked date. - [Commons map](/map/): Interactive diagram of the graph, with the same content as a plain list. - [Catalogue](/catalogue/): Filterable table of learning resources. ## Skills - [practice-coach (garden, not yet reviewed)](/skills/practice-coach.md): A coach for learning to build with AI by improving real work. - [/frame-the-problem (garden, not yet reviewed)](/skills/frame-the-problem.md): Turns a vague complaint or big goal into a specific problem, a desired change and a first thing to try. - [/brainstorm (garden, not yet reviewed)](/skills/brainstorm.md): Shows what's possible for a goal by generating many varied approaches, building on the person's own ideas first. - [/build-brief (garden, not yet reviewed)](/skills/build-brief.md): Plans a small AI build before anything gets built. - [/expert-council (garden, not yet reviewed)](/skills/expert-council.md): Vets an idea, plan, draft or decision from three to five relevant perspectives, each saying what it would do and what worries it, and ends with where they agree and disagree. - [/prove-it (garden, not yet reviewed)](/skills/prove-it.md): Checks an AI result against something real before you trust it or pass it on, using the source, the actual number, or a rerun with one thing changed, and says what evidence would change the answer. - [/stuck-or-pivot (garden, not yet reviewed)](/skills/stuck-or-pivot.md): For when an AI build, prompt or workflow isn't working. - [/make-it-repeatable (garden, not yet reviewed)](/skills/make-it-repeatable.md): Turns something that worked once into something reusable (a saved prompt, template, checklist or personal skill) plus a short share-out of what was built and learned, and can draft it as a contribution to AI Practice Commons. ## Concepts - [Adoption rituals over mandates (garden, not yet reviewed)](/concepts/con-adoption-rituals.md): Teams start using AI through recurring shared habits, such as a weekly AI hour, hack weeks and leaders building things where others can see, more than through orders to use it. - [Agents (garden, not yet reviewed)](/concepts/con-agents.md): AI that works toward a goal over several steps on its own, choosing actions, using tools and checking its progress, rather than giving one reply and waiting. - [AI writing tells (garden, not yet reviewed)](/concepts/con-ai-writing-tells.md): The recognisable habits of AI-generated text, such as stock phrases, tidy groups of three, empty enthusiasm and vague claims, that make writing feel generic and erode readers' trust. - [Allocation economy (garden, not yet reviewed)](/concepts/con-allocation-economy.md): Dan Shipper's forecast that as AI does more of the making, people will be valued less for what they know and more for how well they direct models: a clear goal, saying exactly what is wrong with a result, choosing the right model and knowi… - [Artifacts and canvases (garden, not yet reviewed)](/concepts/con-artifacts.md): A workspace beside the chat where the AI creates a standalone thing you can see, edit and reuse, such as a document, chart, small web page or interactive tool. - [Automation (garden, not yet reviewed)](/concepts/con-automation.md): Setting AI up to handle a repeatable job on its own, triggered by a time or an event like a new form entry, so the work happens without you starting it each time. - [Centaurs and cyborgs (garden, not yet reviewed)](/concepts/con-centaurs-and-cyborgs.md): Ethan Mollick's names for two ways of working well with AI: centaurs split a job, keeping some tasks and handing others to AI, while cyborgs go back and forth with it inside a single task. - [Clarifying questions (garden, not yet reviewed)](/concepts/con-clarifying-questions.md): Asking the AI to question you about the task before it starts, so gaps in your request surface early and you make the decisions the AI would otherwise guess at. - [Cognitive offloading (garden, not yet reviewed)](/concepts/con-cognitive-offloading.md): Handing mental work to a tool. - [Compounding loop (garden, not yet reviewed)](/concepts/con-compound-loop.md): After each piece of AI work, writing down what you learned where the AI will read it next time, such as a line in its instructions, a saved example or a new check, so you don't catch the same mistake twice. - [Context engineering (garden, not yet reviewed)](/concepts/con-context-engineering.md): Choosing and arranging everything the AI can see, such as instructions, examples, files, past messages and connected data, so it has what it needs for the task and little that distracts it. - [Context window (garden, not yet reviewed)](/concepts/con-context-window.md): The amount of material an AI can hold in view during one conversation. - [Custom assistants (garden, not yet reviewed)](/concepts/con-custom-assistants.md): A version of an AI tool set up for one purpose, with its own instructions, files and sometimes tools, that you or others can open and use without setting it up again. - [Custom instructions (garden, not yet reviewed)](/concepts/con-custom-instructions.md): Standing instructions you save once, such as your role, preferred tone or formatting rules, that the AI applies to every conversation so you don't repeat yourself. - [Deep research (garden, not yet reviewed)](/concepts/con-deep-research.md): An AI mode that spends several minutes searching many sources, reading and cross-checking them, and returns a longer cited report on a question. - [Delegating to AI (garden, not yet reviewed)](/concepts/con-delegating-to-ai.md): Deciding whether a task is worth handing to AI, by weighing how long it takes you against how often the AI gets it right and how long asking and checking take, then handing it over with a brief like you'd give a colleague. - [Evals (garden, not yet reviewed)](/concepts/con-evals.md): A small, reusable set of test tasks with known good answers that you run an AI setup against, so you can tell whether it works and whether a change made it better or worse. - [Examples in prompts (garden, not yet reviewed)](/concepts/con-examples-in-prompts.md): Showing the AI one or more samples of what you want, such as a past email or a filled-in table, so it can match the pattern you show it. - [Generative AI (garden, not yet reviewed)](/concepts/con-generative-ai.md): Software that produces new text, images, audio or code in response to a request, by learning patterns from huge amounts of existing material. - [Hallucination (garden, not yet reviewed)](/concepts/con-hallucination.md): When AI states something false or invented, such as a statistic, quote, source or policy detail, with the same confidence as something true. - [Human in the loop (garden, not yet reviewed)](/concepts/con-human-in-the-loop.md): Designing AI work so a person reviews or approves at the points that matter, especially before anything is sent, published, paid or deleted. - [Jagged frontier (garden, not yet reviewed)](/concepts/con-jagged-frontier.md): The uneven line between tasks AI does well and tasks it does badly, which often doesn't match human intuition about what is hard, so you have to test it on your own tasks. - [Large language model (garden, not yet reviewed)](/concepts/con-large-language-model.md): The engine behind AI chat tools: a system trained on vast amounts of text to predict what words should come next, which turns out to be enough to write, summarise, translate and reason in useful ways. - [Memory (garden, not yet reviewed)](/concepts/con-memory.md): A feature that lets an AI carry details about you and your past conversations into new ones, so it can personalise answers without you repeating yourself. - [Model choice (garden, not yet reviewed)](/concepts/con-model-choice.md): Picking which AI model, or which tool, to use for a given job, based on what it is good at, how fast and costly it is, and how much is at stake. - [Model Context Protocol (MCP) (garden, not yet reviewed)](/concepts/con-mcp.md): An open standard for connecting AI tools to other software and data, so a connector built once works in every AI product that supports the standard. - [Multimodal AI (garden, not yet reviewed)](/concepts/con-multimodal.md): AI that can take in or produce more than text, such as photos, screenshots, PDFs, voice, audio and video, so you can show it the problem as well as describe it. - [Onboard AI like a new hire (garden, not yet reviewed)](/concepts/con-onboard-ai-like-a-new-hire.md): Setting up an AI assistant or agent the way you'd bring on a new colleague: a role and ways of working, background on the organisation, its own limited accounts, and more access only as it proves reliable. - [Privacy and data (garden, not yet reviewed)](/concepts/con-privacy-and-data.md): Knowing what information is safe to share with an AI tool, how that tool stores and uses it, and which settings and account types protect you, your organisation and the people you serve. - [Productive friction (garden, not yet reviewed)](/concepts/con-productive-friction.md): Deliberately adding a small amount of effort to working with AI, such as drafting first, predicting before the reveal or asking it to argue back, so you think the problem through before accepting an answer. - [Projects and knowledge (garden, not yet reviewed)](/concepts/con-projects-and-knowledge.md): A workspace where you keep the files, instructions and chats for one body of work together, so every conversation in it starts with the same background. - [Prompt engineering (garden, not yet reviewed)](/concepts/con-prompt-engineering.md): The early practice of finding special wording, templates and tricks to coax better answers from AI. - [Prompt injection (garden, not yet reviewed)](/concepts/con-prompt-injection.md): Anything an AI reads while working, such as an email, web page, calendar invite or document, can contain words written to look like orders. - [Prompting (garden, not yet reviewed)](/concepts/con-prompting.md): Telling an AI what you want in plain words: the task, who it is for, what good looks like and any limits. - [Reasoning models (garden, not yet reviewed)](/concepts/con-reasoning-models.md): Models that work through a problem step by step before answering, trading speed for better results on tasks like analysis, planning, maths and multi-step decisions. - [Scheduled tasks (garden, not yet reviewed)](/concepts/con-scheduled-tasks.md): Asking an AI to run a prompt or job automatically at a set time or on a repeat, such as a Monday news digest or a daily inbox summary, without you starting it each time. - [Skills (garden, not yet reviewed)](/concepts/con-skills.md): Packaged instructions, examples and sometimes small tools for one kind of task, which the AI loads only when that task comes up, so a proven way of working can be reused and shared. - [Smallest useful version (garden, not yet reviewed)](/concepts/con-smallest-useful-version.md): The simplest thing that solves part of your problem, built first and tried for real before adding anything more. - [Solution ladder (garden, not yet reviewed)](/concepts/con-solution-ladder.md): A way to choose how to solve a problem by climbing from the simplest option (paper, a document or sheet, a prompt) toward an agent or an app, and stopping at the lowest rung that works. - [Steering (garden, not yet reviewed)](/concepts/con-steering.md): Directing AI work as it happens: setting the goal and limits, approving one step at a time, correcting course early and cutting anything it does beyond what you need. - [Sycophancy (garden, not yet reviewed)](/concepts/con-sycophancy.md): The tendency of AI to agree with you, praise your ideas and soften bad news, even when you're wrong, because it was trained partly on what people rate highly. - [Tools and connectors (garden, not yet reviewed)](/concepts/con-tools-and-connectors.md): Ways for an AI to do more than write: search the web, read your email or calendar, look in shared drives, or take actions in other apps, using access you grant. - [Verification (garden, not yet reviewed)](/concepts/con-verification.md): Checking what AI gives you against something real, such as the source, the data, an expert or your own knowledge, before you rely on it or pass it on. - [Vibe coding (garden, not yet reviewed)](/concepts/con-vibe-coding.md): Building working software by describing what you want in plain language and letting AI write the code, while you test, steer and decide what it should do. ## Blueprints (project cards) - [Build a style guide so AI drafts sound like you (garden, not yet reviewed)](/projects/prj-ai-style-guide-from-your-samples.md): A one-page style guide, built from your own writing, that you paste into an AI tool before it drafts for you, so you spend less time rewriting its drafts into your voice. - [Ask AI to find where your draft misleads (garden, not yet reviewed)](/projects/prj-ask-ai-where-your-draft-misleads.md): A draft you wrote yourself, revised by you after an AI pointed out every place it overstates, misleads or oversimplifies, with a record of which criticisms you accepted and why. - [Compare AI tools on a task you actually do (garden, not yet reviewed)](/projects/prj-compare-models-on-a-real-task.md): A choice of AI tool for one recurring task, based on a rubric you wrote before seeing any answers rather than habit, rankings or which answer sounded most confident. - [Write a handover brief before you give an agent a real task (garden, not yet reviewed)](/projects/prj-delegation-brief-for-an-agent.md): One task handed to an AI agent with a written brief, a decision about whether it was worth handing over, and a check of the result against the "done" line you wrote in advance. - [Ask AI to play devil's advocate on a team decision (garden, not yet reviewed)](/projects/prj-devils-advocate-for-a-team-decision.md): A team decision that has been questioned before you commit to it, with a short table of the assumptions it rests on and one change or check you'll make because of them. - [Hire and onboard an AI copilot for one piece of work (garden, not yet reviewed)](/projects/prj-onboard-an-ai-copilot.md): An AI project that knows your role, your goals and one live initiative well enough to ask useful questions, push back on a weak plan, and suggest a next step you can judge against your own. - [Let AI ask before it plans (garden, not yet reviewed)](/projects/prj-plan-with-clarifying-questions.md): A plan for a real piece of work that fits your situation on the first or second try, because the details that were in your head made it into the request. - [Teach your AI assistant your team's jargon, one confirmed term at a time (garden, not yet reviewed)](/projects/prj-teach-your-assistant-your-jargon.md): A short glossary of your team's internal terms that the AI found in your own notes, guessed at, and saved only after you confirmed or corrected each one, so later answers stop misreading your shorthand. - [Turn messy notes into a status draft you can check (garden, not yet reviewed)](/projects/prj-weekly-status-from-notes.md): A weekly status update that takes less time to write, points to the note behind every claim, and flags what is missing or contradictory instead of smoothing it over. - [Write a one-page build plan before you ask AI to make an app (garden, not yet reviewed)](/projects/prj-write-a-build-readme-before-you-vibe-code.md): A one-page plan for a small app, with what it must do, what it leaves out and three milestones, that you can hand to any AI app builder so the first version matches what you meant and is easy to check. ## Optional ### Skills from the field (by others, linked not copied) - [academy-guide by Anthropic (Apache-2.0)](https://github.com/anthropics/skills/tree/main/skills/academy-guide): When you ask how to do something with Claude, points you to matching Claude Academy courses, tutorials and use cases. - [discernment-nudge by Anthropic (Apache-2.0)](https://github.com/anthropics/skills/tree/main/skills/discernment-nudge): After a substantive answer you might act on, adds two or three short questions that help you check key facts, probe the reasoning and notice missing context, at most once per conversation. - [skill-creator by Anthropic (Apache-2.0)](https://github.com/anthropics/skills/tree/main/skills/skill-creator): Helps you turn a way of working into your own skill, and test whether it does what you meant. - [ChatPRD plugin (Cursor, Codex, ChatGPT) by ChatPRD (Claire Vo's company) (MIT)](https://github.com/ChatPRD/cursor-plugin): Seven skills for writing a requirements doc from notes or code, scoring a doc on strategy, structure, clarity and completeness, turning a doc into an interactive page, and checking that what was built matches the doc. - [evals-skills by Hamel Husain (companion to the AI Evals for Engineers & PMs course with Shreya Shankar) (MIT)](https://github.com/hamelsmu/evals-skills): For when something you built with AI runs again and again and needs real testing: reviewing failures, writing judge prompts and checking that your checks are trustworthy. - [compound-engineering (plugin) by Every (maintained by Kieran Klaassen and Trevin Chow) (MIT)](https://github.com/EveryInc/compound-engineering-plugin): About 36 skills for AI coding agents, built around a loop: plan, build, review, then write down what you learned so the next job can read it. - [draft-review-kit by Katie Parrott (Every) (MIT)](https://github.com/EveryInc/draft-review-kit): Review skills for a draft you have already written. - [ec-public (consulting workflows) by Every Consulting (MIT)](https://github.com/EveryInc/ec-public): Four skills for everyday client work: a weekly update from material you supply, a proposal from discovery notes, email in a voice you describe, and a review of how a spreadsheet reads to someone new. - [awesome-pm-skills by Udi Menkes (MIT (with an attribution section crediting Lenny's Podcast))](https://github.com/menkesu/awesome-pm-skills): 25 workflows that start from something you already have, such as a pricing page, interview notes or a rough feature idea, and turn it into a spec, a research summary or a test plan. - [no-ai-slop by Peter Yang (MIT)](https://github.com/petergyang/no-ai-slop): Finds the patterns that make writing sound machine-made and suggests edits that keep your meaning and tone. - [human-review by Peter Yang (MIT)](https://github.com/petergyang/human-review): Opens a page or document your AI made in the browser, where you edit the text directly and leave comments as you would in a shared doc. - [fuck-cancer (patient and caregiver brief) by Peter Yang (MIT)](https://github.com/petergyang/fuck-cancer): For someone supporting a person through cancer. - [lenny-skills by Refound AI (MIT)](https://github.com/RefoundAI/lenny-skills): About 76 product and team skills, such as writing a requirements doc, running customer interviews, prioritising a roadmap and giving feedback, each built from methods guests described on Lenny's Podcast and Newsletter. - [Wharton Generative AI Labs prompt library by Ethan Mollick and Lilach Mollick (Wharton Generative AI Labs) (CC-BY-4.0 (the prompts only; other site content is not covered))](https://gail.wharton.upenn.edu/prompt-library/): Seventeen paste-in prompts that set AI up as a coach or role-play partner that asks one question at a time, including a devil's advocate, a team after-action review, a team charter, a meeting facilitator and a tutor. ### Catalogue resources - [How AI assistance impacts the formation of coding skills (Anthropic) (garden, not yet reviewed)](https://www.anthropic.com/research/AI-assistance-coding-skills): A controlled experiment in which people learned a new programming tool with or without AI help. - [AI Capabilities and Limitations (Anthropic) (garden, not yet reviewed)](https://academy.claude.com/courses/ai-capabilities-and-limitations): A plain-language course on how AI chat tools work: why they predict text, why their knowledge stops at a certain date, how much they can hold in mind at once, and how to figure out what went wrong when an answer surprises you. - [AI Fluency: Framework & Foundations (Anthropic) (garden, not yet reviewed)](https://academy.claude.com/courses/ai-fluency-framework-foundations): A free self-paced course that gives you a simple way to think about working with AI: deciding what to hand over, explaining the job clearly, judging what comes back, and taking responsibility for how you use it. - [The AI Fluency Index (Anthropic) (garden, not yet reviewed)](https://academy.claude.com/tutorials/the-ai-fluency-index): Anthropic studied about ten thousand real Claude conversations to see which good habits people show. - [AI Fluency for Nonprofits (Anthropic) (garden, not yet reviewed)](https://academy.claude.com/courses/ai-fluency-for-nonprofits): Applies the AI Fluency habits to nonprofit work such as fundraising, communications, programs and operations. - [Claude 101 (Anthropic) (garden, not yet reviewed)](https://academy.claude.com/courses/claude-101): A guided tour of Claude for people starting to use it at work. - [Anthropic Education Report: How university students use Claude (Anthropic) (garden, not yet reviewed)](https://www.anthropic.com/news/anthropic-education-report-how-university-students-use-claude): A look at how university students use Claude. - [Introduction to Claude Cowork (Anthropic) (garden, not yet reviewed)](https://academy.claude.com/courses/introduction-to-claude-cowork): Shows people who shuffle information between files, apps and tools how to hand multi-step work to Claude's Cowork mode: setting up a workspace, giving it context, and running tasks on a schedule. - [Claude Help Center: Introduction to prompt design (Anthropic) (garden, not yet reviewed)](https://support.claude.com/en/articles/7996853-introduction-to-prompt-design): A short help article on writing requests to Claude. - [Google AI Essentials (Google) (garden, not yet reviewed)](https://grow.google/ai-essentials/): A short certificate course from Google, delivered through Coursera, for people with no AI experience. - [Gemini API prompt design strategies (Google) (garden, not yet reviewed)](https://ai.google.dev/gemini-api/docs/prompting-strategies): Google's guide to designing prompts for Gemini, written for developers but readable by confident nontechnical users. - [Introduction to Generative AI (Google) (garden, not yet reviewed)](https://www.skills.google/course_templates/536): A 45-minute introductory course explaining what generative AI is, what it is used for, and how it differs from older kinds of machine learning. - [Gemini for Workspace prompting guide (PDF) (Google) (garden, not yet reviewed)](https://services.google.com/fh/files/misc/workspace_with_gemini_prompting_guide.pdf): A downloadable guide from Google Workspace on writing prompts for Gemini inside everyday office tools. - [Google Workspace Learning Center (Google) (garden, not yet reviewed)](https://workspace.google.com/learning/content/learning-center): Google's help hub for Workspace apps, including how-to material on using Gemini in Docs, Gmail and other tools. - [Agents Course, Unit 0-1 (what an agent is) (Hugging Face) (garden, not yet reviewed)](https://huggingface.co/learn/agents-course/unit0/introduction): The first units of Hugging Face's free agents course explain what an AI agent is: a model that can plan steps and use tools rather than only answer. - [HuggingChat (try open models) (Hugging Face) (garden, not yet reviewed)](https://huggingface.co/chat): A chat app where you can talk to many open models, either letting it pick one for you or choosing a model yourself. - [LLM Course, Chapter 1: Transformer models (intro) (Hugging Face) (garden, not yet reviewed)](https://huggingface.co/learn/llm-course/chapter1/1): The opening chapter of Hugging Face's free course on large language models, which explains what these models are and what they can do. - [OpenAI Academy: Apply AI at Work track (OpenAI) (garden, not yet reviewed)](https://academy.openai.com/pages/ai-at-work-bcx7td): A hub page collecting OpenAI Academy courses and events on using AI in everyday work, with separate groups for small businesses, nonprofits and government. - [OpenAI Academy: AI Foundations (OpenAI) (garden, not yet reviewed)](https://academy.openai.com/public/courses/ai-foundations-juzjs): OpenAI's beginner course for people who use AI at work, which runs inside ChatGPT and ends with a certificate. - [ChatGPT 101: A Guide to Your AI Super Assistant (recording) (OpenAI) (garden, not yet reviewed)](https://academy.openai.com/public/videos/chatgpt-101-a-guide-to-your-ai-superassistant-recording): A recorded introductory session from OpenAI Academy for people new to ChatGPT. - [OpenAI Academy: Nonprofits club (OpenAI) (garden, not yet reviewed)](https://academy.openai.com/public/clubs/nonprofits-8kc1e/overview): An OpenAI Academy community space for nonprofit staff, with dated events and sessions. - [Prompt engineering best practices for ChatGPT (OpenAI) (garden, not yet reviewed)](https://help.openai.com/en/articles/10032626-prompt-engineering-best-practices-for-chatgpt): OpenAI's help-center article on writing better requests to ChatGPT. - [Generative AI without guardrails can harm learning (PNAS) (garden, not yet reviewed)](https://doi.org/10.1073/pnas.2422633122): A field experiment with about a thousand high-school maths students. - [Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning (UCLA Bjork Lab) (garden, not yet reviewed)](https://bjorklab.psych.ucla.edu/wp-content/uploads/sites/13/2016/04/EBjork_RBjork_2011.pdf): A classic, readable chapter from learning science on why some struggle helps you remember: spacing practice out, mixing topics, testing yourself and producing answers before looking them up. - [Why "Get With the Program" Doesn't Work: Claire Vo on AI That Actually Sticks (Agents of Scale) (Claire Vo) (garden, not yet reviewed)](https://agentsofscale.transistor.fm/23): A 51-minute conversation on Zapier's Agents of Scale podcast about why AI mandates and hiring freezes don't get a team using AI, and what does: regular rituals such as AI Fridays and hack weeks, leaders who build things themselves in the o… - [The AI Agent Glossary (Claire Vo) (garden, not yet reviewed)](https://github.com/ChatPRD/agent-glossary): A community glossary, started from a post by Claire Vo, of agent words such as agent, sandbox, context window, system prompt and MCP. - [How to Turn a One-Off AI Chat Session into a Reusable Workflow for Your Team (Claire Vo) (garden, not yet reviewed)](https://www.chatprd.ai/how-i-ai/workflows/how-to-turn-a-one-off-ai-chat-session-into-a-reusable-workflow-for-your-team): Taken from the How I AI episode on Kai, Stripe's internal AI assistant. - [Claude Cowork 101: How to automate your workday without touching code (How I AI, with JJ Englert) (Claire Vo) (garden, not yet reviewed)](https://www.youtube.com/watch?v=jwGQ9CrqVdA): JJ Englert shows how someone who doesn't code can set up Claude Cowork: a "brain" file that tells Claude about you and your work, an email-writing skill built from your own sent mail, a panel of simulated reviewers that critique one draft… - [How I AI (podcast) (Claire Vo) (garden, not yet reviewed)](https://www.youtube.com/@howiaipodcast): A weekly video show hosted by Claire Vo where guests share their screen and show how they use AI in their actual job: product managers, designers, founders, engineers and parents. - [How I AI workflow guides (Claire Vo) (garden, not yet reviewed)](https://www.chatprd.ai/how-i-ai/workflows): About 290 short step-by-step guides, each taken from a How I AI episode where a guest shows a real AI workflow on screen. - [How to Write a PRD: A Step-by-Step Guide for Product Managers (Claire Vo) (garden, not yet reviewed)](https://www.chatprd.ai/learn/how-to-write-a-prd): A ten-step guide to writing a product requirements document: the problem, who it is for, goals, user stories, priorities, risks and how to get people to agree, with a good and a bad example and three free templates. - [How to Improve Interview Feedback Consistency with a Custom GPT (Claire Vo) (garden, not yet reviewed)](https://www.chatprd.ai/how-i-ai/workflows/how-to-improve-interview-feedback-consistency-with-a-custom-gpt): A beginner guide to a custom GPT that holds your interview rubric plus strong and weak example scorecards, and reviews how well interviewers wrote up their evidence. - [OpenClaw: The complete guide to building, training, and living with your personal AI agent (Claire Vo) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/openclaw-the-complete-guide-to-building): Claire Vo's guest post on setting up OpenClaw, an open-source personal agent you message through apps such as Telegram. - [Claude Cowork for PMs: My Self-Improving Productivity System (How I AI, with Daniel Blum) (Claire Vo) (garden, not yet reviewed)](https://www.chatprd.ai/how-i-ai/claude-cowork-for-pms-my-self-improving-productivity-system): Daniel Blum, a product manager at Melio, shows a Claude Cowork setup that preps his week, writes a morning brief from meeting notes and messages, and asks him to define internal terms it doesn't know before saving them. - [Tradclaw: a household starter kit for OpenClaw (Claire Vo) (garden, not yet reviewed)](https://github.com/ChatPRD/tradclaw): A starter set of files for running a personal AI agent as a family manager: calendar briefs, school email triage, meal plans and shopping lists, home maintenance reminders and stories for kids. - [The Agent That Saved My Brain (Every) (garden, not yet reviewed)](https://every.to/p/the-agent-that-saved-my-brain): Every's head of growth, who says he has no technical background, describes the "command center" he built in Claude Code: a set of notes about the business, connections to the tools he uses, and dozens of small skills for jobs he repeats, s… - [AI Style Guides: How to Help AI Write Like You (Every) (garden, not yet reviewed)](https://every.to/guides/ai-style-guide): A step-by-step guide to writing a style guide that an AI reads before it drafts for you. - [The Knowledge Economy Is Over. Welcome to the Allocation Economy. (Every) (garden, not yet reviewed)](https://every.to/chain-of-thought/the-knowledge-economy-is-over-welcome-to-the-allocation-economy): Dan Shipper argues that as AI takes over more of the making, people get paid less for what they know and more for how well they direct models. - [ChatGPT for Knowledge Work (Every) (garden, not yet reviewed)](https://every.to/guides/codex-for-knowledge-work): A guide to using ChatGPT's agent features for office work, split into tasks you can hand off and tasks you should work through together. - [How to Use Claude Code as a Second Brain (Every) (garden, not yet reviewed)](https://every.to/podcast/how-to-use-claude-code-as-a-thinking-partner): An AI & I episode where Noah Brier shows how he points Claude Code at his folder of notes and tells it he is in "thinking mode", so it asks him questions and finds his past research instead of drafting. - [Compound Engineering (Every) (garden, not yet reviewed)](https://every.to/guides/compound-engineering): Every's long guide to working with AI agents so that each job leaves the next one easier. - [Compound Writing: The Ultimate Guide (Every) (garden, not yet reviewed)](https://every.to/guides/compound-writing): Applies the compound loop to writing: interview, outline, draft, review, finish, then save what you learned so the next piece goes faster. - [The Eight Levels of AI Adoption (Every) (garden, not yet reviewed)](https://every.to/guides/the-eight-levels-of-ai-adoption): Lays out eight levels of handing work to AI, from a chatbot that answers one request, through an agent that checks in before each step and one that works without approvals, up to a manager agent directing others. - [We Gave Every Employee an AI Agent. Here's What Happened. (Every) (garden, not yet reviewed)](https://every.to/podcast/transcript-we-gave-every-employee-an-ai-agent-here-s-what-happened): Every's COO and head of platform talk through giving each staff member their own agent. - [An Executive's Guide to Implementing AI (Every) (garden, not yet reviewed)](https://every.to/guides/an-executive-s-guide-to-implementing-ai): A five-step plan for leaders bringing AI into an organisation: use the tools yourself first, give named champions protected time, pick one frequent and painful workflow, build it with good examples, checks and a named owner, then spread wh… - [Securing an Always-on AI Employee (Every) (garden, not yet reviewed)](https://every.to/guides/securing-an-always-on-ai-employee): How Every limits the risk of an agent that runs all day with real accounts. - [Your Brain on ChatGPT (MIT Media Lab) (garden, not yet reviewed)](https://arxiv.org/abs/2506.08872): An MIT Media Lab study that measured brain activity while people wrote essays with ChatGPT, a search engine or nothing. - [The Impact of Generative AI on Critical Thinking (CHI) (garden, not yet reviewed)](https://doi.org/10.1145/3706598.3713778): A survey of 319 knowledge workers about when they think critically while using AI. - [Not all AI agents are created equal (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/not-all-ai-agents-are-created-equal): Sorts agent ideas into three kinds before you compare effort or value: fixed step-by-step automations where AI fills in content at set points, agents that choose their own next step from a set of tools, and networks of agents owned by diff… - [An AI glossary (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/an-ai-glossary): Plain-language definitions of about 20 AI terms, from model, token and training to RAG, MCP, agents, evals and hallucination, written for people in meetings where these words fly past. - [A guide to AI prototyping for product managers (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/a-guide-to-ai-prototyping-for-product): Colin Matthews compares three kinds of tool for turning an idea into a clickable prototype: chat assistants, browser builders such as Bolt, v0, Lovable and Replit, and coding assistants such as Cursor, and says when to use each. - [Build your personal AI copilot (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/build-your-personal-ai-copilot): Tal Raviv's free guide to using an AI project as a long-term thinking partner at work. - [Lenny's Data: free public starter pack (Lenny) (garden, not yet reviewed)](https://github.com/LennysNewsletter/lennys-newsletterpodcastdata): A free sample of Lenny's archive as plain Markdown: 10 newsletter posts and 50 podcast transcripts, with an index of titles, dates and guests. - [Beyond vibe checks: A PM's complete guide to evals (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/beyond-vibe-checks-a-pms-complete): Aman Khan of Arize AI explains how to test AI output on purpose rather than by feel. - [Everyone should be using Claude Code more (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/everyone-should-be-using-claude-code): Lenny walks through installing Claude Code and then collects 50 ways people outside engineering use it: cleaning up messy invoice files, organising folders, synthesising customer-call transcripts, drafting a job description, building slide… - [This week on How I AI: Claude Cowork tutorial for non-engineers + Build your own Slack inbox (for $0) (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/this-week-on-how-i-ai-claude-cowork): A free weekly roundup of two How I AI episodes: JJ Englert's Claude Cowork tutorial for people who don't code, and Yash Tekriwal's custom Slack digest, which cut 100 to 150 daily notifications to about 30 that need attention. - [The non-technical PM's guide to building with Cursor (Lenny's Podcast, with Zevi Arnovitz) (Lenny) (garden, not yet reviewed)](https://www.youtube.com/watch?v=1em64iUFt3U): Zevi Arnovitz, a product manager at Meta who doesn't write code, shows how he builds working products with Cursor and Claude. - [How to build your PM second brain with ChatGPT (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/how-to-build-your-pm-second-brain): Amir Klein, a product manager at monday.com, shows how he set up a ChatGPT Project as a second brain for a new initiative: write instructions that make it challenge him, upload everything he had (docs, decks, exported Slack threads) as fil… - [What people are vibe coding (and actually using) (Lenny) (garden, not yet reviewed)](https://www.lennysnewsletter.com/p/what-people-are-vibe-coding-and-actually): More than 50 small tools that readers built with Replit, Lovable, Bolt, Cursor and similar and still use: a carb counter, a chore app for kids, a bill splitter, meeting-prep automation, a tracker of wins at work. - [A Conversation with Dr. Vivienne Ming, author of Robot-Proof (Ming) (garden, not yet reviewed)](https://www.fcatalyst.com/trends-and-signals/a-conversation-with-dr-vivienne-ming-author-of-robot-proof): A short written Q&A from Fidelity's research centre. - [A top researcher says a new divide is emerging in AI use, and most people are on the losing side (Ming) (garden, not yet reviewed)](https://www.aol.com/articles/top-researcher-says-divide-emerging-143334376.html): A news report on Ming's forecasting experiment, in which teams predicted real events with or without AI and were compared against Polymarket. - [Vivienne Ming on hybrid collective intelligence, building cyborgs, meta-uncertainty, and the unknown infinite (Ming) (garden, not yet reviewed)](https://humansplus.ai/podcast/vivienne-ming-hybrid-collective-intelligence-building-cyborgs-meta-uncertainty-unknown-infinite-ac-ep13/): A long interview on the Humans + AI podcast with a full transcript. - [In Conversation... with Vivienne Ming (Ming) (garden, not yet reviewed)](https://www.lewissilkin.com/en/our-thinking/future-of-work-hub/insights/2024/10/02/in-conversation-with-vivienne-ming): A 34-minute interview from a law firm's future-of-work series, with a written transcript. - [Robot-Proof: When Machines Have All the Answers, Build Better People (Ming) (garden, not yet reviewed)](https://www.wiley.com/en-us/Robot-Proof%3A+When+Machines+Have+All+the+Answers%2C+Build+Better+People-p-9781394397808): Vivienne Ming's 2026 book about people more than machines. - [The AI That Asks Better Questions: Dr. Vivienne Ming at SIM 2025 (Ming) (garden, not yet reviewed)](https://citynomads.com/the-ai-that-asks-better-questions-dr-vivienne-ming-at-sim-2025/): A write-up of Ming's 2025 keynote in Porto. - [15 Times to use AI, and 5 Not to (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/15-times-to-use-ai-and-5-not-to): A checklist for deciding whether a task suits AI. - [Against "Brain Damage" (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/against-brain-damage): Mollick pushes back on headlines claiming AI harms our brains and says the danger is thoughtless use. - [Giving your AI a Job Interview (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/giving-your-ai-a-job-interview): Explains why public benchmark scores tell you little about whether a model suits your work, and suggests testing models the way you'd interview a job candidate: give them realistic tasks from your field, grade the answers blind and repeat… - [Assigning AI: Seven Approaches for Students, with Prompts (Mollick) (garden, not yet reviewed)](https://arxiv.org/abs/2306.10052): Describes seven roles you can give an AI, such as mentor, tutor, coach, teammate, student, simulator and tool, each with example prompts and the risks to watch for. - [Centaurs and Cyborgs on the Jagged Frontier (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/centaurs-and-cyborgs-on-the-jagged): Mollick's plain-language account of a 2023 experiment with 758 Boston Consulting Group consultants. - [Choosing to Stay Human (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/choosing-to-stay-human): Ethan Mollick argues that the risk lies in using AI on autopilot, and pulls together studies showing that skipping effort means skipping learning. - [Co-Intelligence: Living and Working with AI (Mollick) (garden, not yet reviewed)](https://www.penguinrandomhouse.com/books/741805/co-intelligence-by-ethan-mollick/): Mollick's 2024 book for general readers on working alongside AI as a coworker, tutor, coach and creative partner. - [The Cybernetic Teammate (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/the-cybernetic-teammate): Reports a randomised trial with 776 Procter & Gamble professionals working on real product development tasks. - [Wharton Generative AI Labs Prompt Library (Mollick) (garden, not yet reviewed)](https://gail.wharton.upenn.edu/prompt-library/): Seventeen tested prompts you paste into a chat tool, each of which sets the AI up as a coach, tutor or role-play partner that asks one question at a time. - [Getting started with AI: Good enough prompting (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/getting-started-with-ai-good-enough): A short starter guide for people who feel they need to learn prompt engineering first. - [Instructors as Innovators: A future-focused approach to new AI learning opportunities, with prompts (Mollick) (garden, not yet reviewed)](https://arxiv.org/abs/2407.05181): A 76-page guide with ready-to-use prompts for AI exercises: simulations, mentoring, coaching and co-creation, each with notes on how to run it, how to assess it and what can go wrong. - [Making AI Work: Leadership, Lab, and Crowd (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/making-ai-work-leadership-lab-and): A plan for bringing AI into an organisation through three groups. - [Management as AI superpower (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/management-as-ai-superpower): Argues that working with AI agents mostly uses ordinary management skills: delegating, giving clear instructions and judging the work. - [An opinionated guide to which AI to use to do stuff (Summer 2026 Edition) (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/an-opinionated-guide-to-which-ai-b22): Mollick's mid-2026 buyer's guide for people who aren't developers. - [On Working with Wizards (Mollick) (garden, not yet reviewed)](https://www.oneusefulthing.org/p/on-working-with-wizards): As AI produces larger pieces of finished work, you can see less of how it got there, and checking every detail can take as long as doing the job. - [12 Rules to Vibe Code Without Frustration (Peter Yang) (garden, not yet reviewed)](https://creatoreconomy.so/p/12-rules-to-vibe-code-without-frustration): Peter Yang's tips from more than 50 hours of building apps with AI as someone who doesn't code. - [20 Tips to Master Claude Code in 35 Minutes (Build a Real App) (Peter Yang) (garden, not yet reviewed)](https://creatoreconomy.so/p/20-tips-to-master-claude-code-in-35-min-build-an-app): A product manager's tour of Claude Code while building an app. - [Introducing My New AI Learning Path and Co-Pilot (Peter Yang) (garden, not yet reviewed)](https://creatoreconomy.so/p/introducing-my-new-ai-learning-path-and-co-pilot): A starting map to Peter Yang's tutorials, sorted into six tracks: a personal AI assistant, career, saving time at work, prototypes and apps, workflows and agents, and building AI products. - [Behind the Craft — Transcripts (Peter Yang) (garden, not yet reviewed)](https://github.com/petergyang/behind-the-craft-transcripts): Full transcripts of Peter Yang's YouTube tutorials and interviews (181 episodes when checked), each tagged with title, guest, date and topics, plus a topic index. - [Behind the Craft (YouTube channel) (Peter Yang) (garden, not yet reviewed)](https://www.youtube.com/@PeterYangYT): Peter Yang's channel of practical AI tutorials and interviews with people who build AI products. - [ChatGPT vs Claude vs Gemini: The Best AI Model for Each Use Case in 2025 (Peter Yang) (garden, not yet reviewed)](https://creatoreconomy.so/p/chatgpt-vs-claude-vs-gemini-the-best-ai-model-for-each-use-case-2025): Peter Yang gives the same tasks to three AI assistants, such as editing his own draft from his writing samples and researching a market, and says which one he would use for each kind of work. - [Use My /Human-Review Skill to Edit HTML and Markdown Files Visually (Peter Yang) (garden, not yet reviewed)](https://creatoreconomy.so/p/use-my-human-review-skill-to-edit-html-markdown-visually): Introduces a free skill that opens a document or web page your AI made in the browser, where you edit text directly and leave comments as you would in a shared doc, then send all of it to the agent in one batch. - [n8n Tutorial: Build an AI Workflow to Extract Top Insights From All Your Newsletter Emails (Peter Yang) (garden, not yet reviewed)](https://creatoreconomy.so/p/n8n-tutorial-build-an-ai-workflow-to-get-a-weekly-digest-newsletter-emails): Builds a weekly automation in n8n that collects every email you have labelled as a newsletter, passes them to an AI model with a summary prompt and emails you one digest on Sunday morning. - [Build Beautiful Slides with Claude Code in 12 Minutes (12 Formats + 3 Templates) (Peter Yang) (garden, not yet reviewed)](https://creatoreconomy.so/p/build-beautiful-slides-with-claude-code-in-12-min): Shows a slide-making skill that takes a rough outline, asks you questions about the content, looks up gaps, builds the deck as a single web page and then screenshots each slide to fix its own layout problems. - [How to be more ambitious: A tutorial for humans with models (Felix Rieseberg) (garden, not yet reviewed)](https://felixrieseberg.com/how-to-be-more-ambitious-a-tutorial-for-humans-with-models/): A practical essay on asking more of AI. - [AI Should Challenge, Not Obey (ACM) (garden, not yet reviewed)](https://arxiv.org/abs/2411.02263): Advait Sarkar argues that AI should sometimes act as a provocateur, offering counterarguments, critiques and questions instead of finishing your work for you. - [Perplexity Academy (Perplexity) (garden, not yet reviewed)](https://www.perplexity.ai/hub/academy): A library of short video lessons, a few minutes each, on using Perplexity's search and its Computer feature for multi-step work. - [Perplexity at Work Guide (Perplexity) (garden, not yet reviewed)](https://www.perplexity.ai/hub/guides/perplexity-at-work): A downloadable guide on using Perplexity in daily work, organised around cutting distractions, extending what you can research and write, and reaching concrete outcomes. - [Evaluating AI for high-stakes work (Perplexity) (garden, not yet reviewed)](https://www.perplexity.ai/hub/guides/evaluating-ai-for-high-stakes-work): A guide for teams in finance, legal and research who must check AI work before it shapes a decision. - [Search vs Computer (Perplexity) (garden, not yet reviewed)](https://www.perplexity.ai/hub/academy/search-vs-computer): A short lesson on when a quick cited answer is enough and when to hand over a multi-step job that pulls from several tools, builds a report and can run on a schedule. - [Grok Bot use cases (xAI) (garden, not yet reviewed)](https://docs.x.ai/grok-bot/use-cases): A page of example roles for xAI's Grok Bot, which gathers and prepares work across connected tools and hands it back for review, with outside actions held for your approval. - [Welcome to Grok (consumer docs) (xAI) (garden, not yet reviewed)](https://docs.x.ai/grok/overview): xAI's overview of the Grok assistant and its features.