I used an AI assistant to draft a volunteer management handbook for an arts organization. I wanted a complete starting document I could review and adapt, rather than building one from scratch. 1. I asked the AI to draft a volunteer handbook for the arts organization. 2. It pulled information from the organization's website. 3. It referenced relevant state laws. 4. It included the organization's current volunteer opportunities. 5. It reflected the organization's DEI policies. 6. It took the needs of the Board of Directors into account. 7. It wrote job descriptions and training opportunities for each volunteer role. The output was an 18-page volunteer management handbook with job descriptions and training for each role. The document is fabulous.

My Windows File History backup stopped working, and after a lot of troubleshooting I still couldn't get it running reliably. I wanted to keep using one external USB drive, so I asked ChatGPT to help me design a replacement. The result is a two-part automated system: FreeFileSync handles my everyday files, and a PowerShell script handles my roughly 12 GB Outlook archive. It's for anyone comfortable with Windows who has no programming background. 1. ChatGPT walked me through setting up FreeFileSync to mirror my important documents and folders to the external drive, using several source and destination pairs. 2. We created backup batch files for those pairs. 3. I configured RealTimeSync to watch for changes and trigger the backups automatically. 4. We added exclusion filters so the Outlook PST file stays out of continuous sync, since Outlook is normally running and the file needs special handling. 5. ChatGPT wrote a PowerShell script to back up the Outlook archive separately. 6. Windows Task Scheduler runs the script three times a day to check whether a backup is due, using a minimum 14-day interval. 7. The script detects changes so it skips unnecessary backups. 8. When a backup is due, it asks permission to close Outlook, copies the archive, and reopens Outlook automatically. 9. It verifies the copied archive with a SHA-256 checksum. 10. It keeps two rolling versions, Current and Previous, and logs errors. It also has safeguards against failed copies and disconnected drives. 11. ChatGPT helped me test the setup in isolation, verifying three backup rotation scenarios without touching my real Outlook data. What it produces: a configured, automated backup system on one external drive. Initial tests have passed, and the first full production Outlook backup is still pending.

I trade gold futures (GC and MGC) during the New York session. I wrote this workflow to keep my process consistent before I scale to multiple accounts. It is for traders who want one market, one playbook, one risk framework and one journal. I analyze in TradingView and execute in Tradovate. 1. From 60 to 45 minutes before the session, I review the economic calendar, USD, Treasury yields and gold movement. 2. From 45 to 15 minutes before, I analyze D1/H4 and H1/M15 and mark key liquidity levels: previous-day high and low, Asia and London highs and lows, and major support and resistance. 3. I set my bias (bullish, bearish or neutral) and my daily risk limits. 4. At 9:30 AM ET I observe the US cash-market open. 5. From 9:35 to 11:00 AM ET I execute only confirmed setups, either Setup A (liquidity sweep reversal) or Setup B (trend continuation). I skip trades when structure is unclear, price is sideways, there is no entry confirmation, risk cannot be calculated, or trading restrictions apply. 6. I size each position with this formula, always rounded down: Contracts = Dollar risk budget ÷ (Stop ticks × Tick value + cost allowance). I risk 0.25% of my defined risk budget or less per trade. 7. After the session, I journal every trade, review P&L and evaluate performance. My risk limits: I stop after two planned losses or the firm's lower limit, take a maximum of 3 attempts per session, hold one gold trade idea at a time, and aim for at least 2:1 planned R:R, subject to testing. After any rule violation I stop and review. I never widen stops or increase size to recover losses. What it produces: a repeatable routine, a journal of 30 to 50 trades before I expand, and weekly measures of net profit after fees, win rate, expectancy, maximum drawdown and rule adherence.

I use Instinct, an AI assistant I text or send voice notes to over iMessage, to manage personal to-dos that used to live in my head or in a pile of iPhone alarms. Many tasks come up when I’m away from my desk, so I used to forget them. Alarms didn’t help much either: they went off once, I swiped them away, and the task was gone. Now, when something comes up, I send Instinct a quick voice note. I don’t need to open another app or type. For example: “Remind me tomorrow at 10 AM to respond to book my next dental appointment, and follow up until I say it's done.” Instinct turns the voice note into a clean reminder with a time attached. If I don’t reply “done,” it checks back in and keeps following up until I close out the task. I use the same thread for recurring tasks, partner follow-ups, ad hoc tasks, and advertising to-dos, so everything I owe people lives in one place. Capturing a task takes a few seconds, which means I actually do it. The built-in follow-up keeps things from slipping just because I was busy when the reminder went off, and the workflow lives in iMessage, where I already am throughout the day. Try sending one voice note today with a real task and end it with “follow up until I say it's done.” Step-by-step: 1. When a task comes up, send Instinct a voice note in iMessage. 2. Include the task, when to be reminded, and whether Instinct should follow up until you confirm it is done. 3. Let Instinct turn the voice note into a scheduled reminder. 4. Reply “done” when you complete the task. 5. If you do not reply “done,” let Instinct check back in and continue following up until you close it out. 6. Keep recurring tasks, partner follow-ups, ad hoc tasks, and advertising to-dos in the same iMessage thread.

I built a recovery-aware cycling coach in Claude for myself, a 60-year-old masters rider. It reads my training and recovery data every morning and tells me whether to go hard, go easy, or rest. It is for masters athletes who recover more slowly than they used to and want a coach that plans around that. 1. I set up a Claude Project with a detailed coaching prompt. It covers my role for Claude, my current training plan, my zones and FTP, and the lessons from my own data about recovering at 60: 72 hours between hard efforts, three weeks to recover from a big multi-day block, and missed hard sessions get skipped, not made up. "You are a cycling coach, exercise physiologist, and sports nutritionist for a 60-year-old masters athlete. Prioritize research on athletes aged 45–65. Cite a study for every physiological claim, or don't make it. Plan around slower masters recovery. Refer anything medical to my doctor. Every review ends with interview questions." 2. I connected Strava through MCP so Claude reads my rides, runs, and lifts directly. 3. I connected Garmin's Personal API through an MCP server so Claude reads overnight HRV, resting heart rate, sleep, and Body Battery. 4. A scheduled task on my Mac runs at 6:45 every morning. It pulls Garmin, Strava, and MyFitnessPal data and rewrites my "Training Command Center" dashboard with a readiness call (go hard, easy, or rest) and coaching notes. My phone gets a link. 5. I check in by chat daily and after big sessions. Claude replies with a verdict, findings with numbers, the exact next session (watts, pace, or sets), and two to four questions about what the data can't show, like motivation, sleep, and appetite. 6. My answers outrank the watch. Readiness is ranked by motivation, then sleep, then appetite, with Garmin data breaking ties, because my HRV did not turn out to predict how well I ride. 7. Noteworthy sessions go into a running "Living Doc" in the Project, so the coach keeps what each season taught. The result is a daily readiness call and a specific next session, built around how I actually recover at 60.

I built an interactive web app to track a family member’s blood pressure four times a day. She is monitoring how her readings change throughout the day and how they respond when her doctor adjusts her medications. The app displays graphs and bar charts organized by date, time of day, and medication. This makes it easier to see how her blood pressure varies at different times and with different medications. She can upload readings by taking a picture of the blood pressure monitor screen or by entering the data into text fields. Selected data can be downloaded as a spreadsheet. The app works on her phone and laptop through a link hosted on Claude.ai. Step-by-step: 1. I built the app as a Claude artifact with a built-in database. 2. I hosted the web page on Claude.ai so it could be accessed from her phone and laptop through a link. 3. I connected the page to a database so readings persist and sync across devices. 4. I added photo input so she can take a picture of the monitor screen and have Claude read the numbers. She confirms the readings before they are saved. 5. I added table views, graphs, bar charts, and filters for weekly, monthly, or custom date ranges. 6. I added text fields for manually entering readings. 7. I added an export button for a CSV or printable report that she can take to the doctor.

I used Claude Code to build a scenario-testing financial model for ANT, based on Anthropic's S-1 filing. It is for anyone who wants to see how different assumptions change the projections, rather than reading a static spreadsheet. 1. I pointed Claude Code at Anthropic's S-1 filing and had it extract the historical financials. 2. From that data, it built a complete set of income statements, balance sheets and cash flow statements. 3. I then had Claude Code build an HTML front-end for the model. 4. On the front-end, I added sliders to control the key variables, along with graphical displays that show the projections under each scenario. 5. I deployed the system on a website and wrote up the process on my Substack: https://onabetternote.substack.com/p/build-your-own-ant-financial-model The result is a three-statement model with an interactive interface where I can dial in different scenarios and see the projections change.

Every Friday at 10 a.m. ET, a scheduled Claude agent refreshes a living "Hevy Workout Log" document with that week's strength-training data, body measurements, and a progress photo. The automation runs fully unattended, with no human in the loop after the owner sets it up. The agent pulls information from three independent systems: the Hevy fitness-tracking API, a Google Drive photo folder, and the living document. It reconciles those sources against a date range, determines what's new versus already recorded, and writes structured, formatted updates in the correct places. This requires more judgment than a simple script or webhook. The scheduled task starts a fresh agent session every Friday at 10 a.m. ET with the full task specification and API key included in the prompt. The Hevy API is the source of truth for workouts from `GET /v1/workouts` and body measurements from `GET /v1/body_measurements`; both endpoints are paginated. Google Drive holds the "Progress Pics" folder, which the agent searches by file timestamp. Claude Docs, through a Claude Artifact, is the living document: it has one tab for each weekday and a Weekly Summary tab, and the agent edits it in place with structured operations rather than full rewrites. To embed a photo, the agent downloads the Drive file locally, uploads it to the Claude Artifact asset store, converts it to a document-native blob, and references it as a Markdown image. There is no custom backend, database, or glue code. The agent calls each system's API directly and keeps the orchestration logic in its reasoning for that run. Step-by-step: 1. Read the document's current state and find the most recent week already recorded in the Weekly Summary tab. 2. Call the Hevy API and determine which Monday–Friday week has just ended, making sure today's date is past that week's Friday. 3. Check whether that week's row is already in the Weekly Summary table, fully populated, and contains the latest matching photo. If everything is already recorded, there may be nothing to update, but the agent still checks for a newer photo that may have appeared in Drive after the data was written. 4. Pull per-set details for every workout that week, including the exercise, weight, and reps. Compute total daily lifted volume by summing weight × reps for each set on each day. 5. Search the Google Drive photos folder for images timestamped within that Monday–Sunday window. Select the single most recent image, replacing any older photo already embedded for that week. 6. Pull body measurements from the same week and calculate weekly averages for weight and body fat, along with the week's waist reading. Convert units as needed from kilograms to pounds and centimeters to inches. 7. Update the Weekly Summary tab with one row per week, including daily volumes, weekly averages, and the progress photo. 8. Update each weekday tab with a dated section containing the workout name, a per-exercise set-by-set table, and a Total row for summed volume and reps. If no workout was logged, add a note instead. 9. Send the human a plain-language summary every time, describing what changed or confirming that nothing changed, because nobody is watching the scheduled run live. The workflow is idempotent without a database. There is no job-tracking table; on each run, the agent reads the live document, compares dates, and checks whether cells are filled before deciding what to write. Progress photos are not tagged by week, so the agent matches them only by whether their Drive timestamp falls within the Monday–Sunday window. It explicitly selects the most recent match, allowing a better photo taken later in the week to replace an earlier embedded photo. The workflow also handles the multi-hop process required to embed an image in a live document: Drive file to local download, artifact asset upload, document-native blob conversion, and Markdown image reference. If the Hevy API key is revoked, the task stops cleanly and tells the owner to generate a new key and provide it, rather than failing silently or looping. Because the workflow runs on a schedule without anyone watching, its specification requires a notification every time, including when nothing changed. This keeps the automation auditable even when it performs no updates.

When I create video content, I pull imagery from many different sources, so my editing folder often ends up with photos that may or may not work inside my editing program. For a handful of photos, converting them manually is fine. For larger projects with dozens of images, it eats into my time and focus. I wanted something simple, lightweight on compute power, that could batch convert photos. 1. I asked Claude to create an app that runs on my local machine and converts photos into formats I know will work in my video editor. 2. I asked for it to be able to batch convert photos. 3. Claude returned an app that worked exactly the way I needed. 4. I asked Claude to install it on my computer. It was installed and working within seconds. 5. I use the app to batch convert my photo folders before bringing them into the editor. There are plenty of traditional ways to do this, but with Claude I got a solution that fit my workflow perfectly, and large image batches no longer slow my editing down.

At some point, my newsletters stopped feeling like reading and started feeling like debt. The information was good; I just couldn’t keep up. I wanted a way to turn that backlog back into something useful. So I built The Daily Nexus, a private tool that reads newsletters from a Gmail label and creates two editions: a podcast I can listen to and a separately written, two-page newspaper I can scan. It runs on demand or on a schedule, and it can publish the audio to a private RSS feed for Apple Podcasts. The project also became a hands-on experiment in building with coding agents. Claude Code and Codex helped me implement features, troubleshoot failures, review the design, and tighten security. The stack includes Python, the Gmail API, Antigravity, Kokoro, FFmpeg, Firebase, Cloudflare Workers, and GitHub Actions. The carousel shows the rest of the flow. It started as a personal tool, but I’m sharing the template for anyone who wants to adapt the idea. Each deployment uses its own accounts and credentials, and the design aims to avoid additional API costs by using an existing AI subscription and available free tiers. GitHub Repo Template: https://lnkd.in/eYceS4KR Step-by-step: 1. I label the newsletters I want to process in Gmail. 2. I run The Daily Nexus on demand or on a schedule so it can read the newsletters from that Gmail label. 3. The tool creates a podcast edition and a separately written, two-page newspaper edition. 4. I listen to the podcast or scan the newspaper, depending on how I want to catch up. 5. When needed, the audio is published to a private RSS feed for Apple Podcasts. 6. I use Claude Code and Codex to implement features, troubleshoot failures, review the design, and tighten security. 7. Each deployment uses its own accounts and credentials, with Python, the Gmail API, Antigravity, Kokoro, FFmpeg, Firebase, Cloudflare Workers, and GitHub Actions supporting the workflow.

I built a live camera bubble that floats over my slides during Facebook Live presentations. Because Facebook Live does not let me switch between my camera and a shared screen, I wanted my face to stay visible while I presented. This is useful for anyone who streams presentations to a Facebook Group and wants viewers to see them speaking. Claude Code built the working app within about five minutes. I made one tweak to the camera bubble’s shape, and it was then a finished, usable product. When I’m ready to go live in the Facebook Group, I click the camera bubble app on my local desktop. It uses my laptop camera, and I can drag the bubble around the screen and choose the circle size that suits whatever I’m sharing. Before I go live, I select "share entire screen" in the Facebook Live settings. Viewers then see my slides or screen with my floating camera bubble, so they can see me speaking as I present. The result is a live stream that shows my slides and my face together, and it took about five minutes to build. Step-by-step: 1. I gave Claude Code the challenge of building a live camera bubble for Facebook Live presentations. 2. Claude Code built the working app within about five minutes. 3. I made one tweak to the shape of the camera bubble, creating a finished, usable product. 4. When I’m ready to go live in the Facebook Group, I click the camera bubble app on my local desktop and use my laptop camera. 5. I drag the bubble around the screen and choose the circle size that suits what I’m sharing. 6. In the Facebook Live settings, I select "share entire screen." 7. I present my slides or screen while viewers see my floating camera bubble and my face together.

Last year my daughter went to camp and wanted daily updates on her favorite TV show, which ran every night. The camp only accepts letters, and I did not want to handwrite them every night, so I built a workflow to send them for me. It works for anyone who needs to send physical mail on a regular schedule. Through a website, I used Make.com to build the workflow: 1. It starts at a specific time each day. 2. It asks Claude for a recap of the previous night's episode. 3. The recap is fed into the Make.com workflow. 4. Make.com connects to Lob through an API. 5. I submitted through Lob to create the letter. Every late morning I see the new letter go out, and my daughter gets the episode recap a few days later. Recently I used a very similar workflow to build websites that send letters for people who want to cancel gym memberships or send landlords an email to end a rental. Both are situations where only physical letters are accepted. Connecting these tools took a little time, but it opened up a lot of possibilities, as long as someone still needs a physical letter.

I'm becoming GrokBot-pilled. I run my company on top of what's happening on X. I used to check everything manually: scanning for new workflows, searching brand mentions, watching what readers complained about, and trying to remember who I had just met at events. I've now handed all four jobs to GrokBot routines. The combined setup took under an hour, and the routines save me a couple of hours of manual checking every day. Every prompt below is read-only by design: the agent watches, labels, and drafts, but never posts, replies, texts, or emails anyone. Paste each prompt as-is, then replace the bracketed parts with your own channels, brand, and time zone. Routine 1: Find real AI workflows worth stealing and filters out the hype threads. Every weekday at 9:15am in my time zone, search X and the web for new AI workflows people are actually running. Keep only cases that name the old task, the tool, and the result: time saved, a step removed, or a handoff that used to need a person. Drop product announcements, “10 prompts” threads, and anything the author has not used more than once. Return at most 3. For each, give the link, the job it replaces, what I would need connected, and whether it fits my work. Send it to me on Slack. Do not post or reply anywhere else. Save this as a routine. If nothing qualifies, send one line saying so. Routine 2: Labels every mention of my name or company and delivers the results. Every weekday at 9:00am, search X for mentions in the last 24 hours of [your brand, @handle, or your own name]. For each hit, give the author, link, a one-line summary, and whether it is praise, complaint, question, or neutral. Do not reply to anyone. Post the digest in [your Slack channel] and tag me. Save this as a routine. If a source is unavailable, say which one. (Note: works just as well for a personal brand or a job search as it does for a company) Routine 3: Catches people actively looking for what I do. Every 4 hours, search X for people evaluating [your category]. Catch three things: reviews or complaints about [category]; posts asking which [category] to use; and complaints about the problem [your product] solves. Quote the post, link it, name the account, and label it review, shopping, or complaint. Skip anything older than 24 hours, and skip [category] accounts promoting themselves. Post to [your Slack channel] only if there is at least one new hit you haven’t already shared. Do not reply to or interact with anyone. Save this as a routine. If the search fails, report the failure in Slack. Routine 4: Keeps the people I meet at events from disappearing into my contacts app. Every weekday at 8:30am in my time zone, read my address book and list contacts added in the last 14 days. For each, show name, date added, phone, email, and any note. Ask me what we talked about before you write anything. Once I reply, draft a follow-up task for each person in [your task tool] with the context attached. Never text, email, or call anyone. Save this as a routine. If the address book is unavailable, say so and do not use old data. Step-by-step: 1. Paste each routine prompt as-is. 2. Replace the bracketed fields with your own channels, brand, category, product, task tool, and time zone. 3. Save each prompt as a GrokBot routine on its stated schedule. 4. Review the read-only outputs in Slack and respond to Routine 4 when it asks what you discussed with a new contact. 5. Keep the routines limited to watching, labeling, and drafting; they should never post, reply, text, email, or call anyone.

I’m 78, and I run a homestead, a home server, and a senior center, with dozens of jobs underway at once. Most jobs get done a piece at a time, while to-do apps only track whether something is “done” or “not done.” So I built Taskmaster: a plain-text ledger inside a Claude Project. I first tried building a large custom app, then scrapped it. One of my own rules is, “Build the ugly version first.” In the field, I speak notes into the Claude app on my phone. Claude saves my exact words, updates each job’s progress, and asks me one question at a time when something is unclear. When I ask, “What’s next?” it considers deadlines, weather, and my energy level, then gives me one to three things to do. Anything tied to a specific time goes on my Google Calendar. Home Assistant then reads the reminder aloud through my hearing aids. Today, it checked the forecast and scheduled my paint job for the one day this week that stays above 50°F overnight. The machine carries the complexity, so I don’t have to. Step-by-step: 1. I keep each ongoing job in a plain-text ledger inside a Claude Project. 2. When I’m in the field, I speak notes into the Claude app on my phone. 3. Claude saves my exact words, updates each job’s progress, and asks one question at a time when something is unclear. 4. When I ask, “What’s next?” Claude weighs deadlines, weather, and my energy, then gives me one to three tasks. 5. I put anything with a specific time on my Google Calendar. 6. Home Assistant reads calendar reminders aloud through my hearing aids. 7. For example, Taskmaster checked the forecast and scheduled my paint job for the one day this week that stays above 50°F overnight.

I built a quarterly “AI Audit” of my Claude history, adapted from The Rundown’s AI Audit prompt and turned into a repeating loop. Each quarter, Claude reads every conversation I’ve had and reports patterns I can’t see myself: where my stated priorities differ from where my attention actually goes, which decisions I keep reopening, and connections across work, money, travel, and personal life. It then rewrites its own instructions for working with me based on what it found. I request my Claude data export, and the run finds the export email in Gmail, downloads it through Claude in Chrome, and unzips it on my machine through the desktop app. I never touch a zip file. The conversations are condensed and split by month, and Sonnet subagents pull structured notes from each month. An Opus subagent combines the monthly notes into a quarterly report under strict evidence rules: every finding cites at least three conversations across at least two months, states the strongest counter-evidence, and labels each claim as OBSERVED or INFERRED. When the evidence is thin, it says “insufficient evidence” instead of padding. Claude writes and maintains a Google Doc with two tabs. Reports contains one report per quarter with footnoted citations. Progress compares the quarters and includes two sections I added to the original prompt: Stop / Start / Continue / Change, which generates concrete behavior changes for the next quarter, and a Cross-domain pass, which identifies patterns spanning work, money, body, and personal life that I would not connect on my own. I push back on findings I disagree with. Those corrections go into the doc and a versioned prompt, now v1.1, so the next run does not repeat the mistake. Claude then takes the latest audit and its stored memory about me and proposes rewritten project instructions. The first rewrite removed stale facts and added rules to check memory, docs, and connected apps before saying it does not have something; move long-running threads into a maintained doc because one tracking chat had grown to 434 messages; and name the number that would settle an open decision because the audit showed I close decisions once a number exists and keep circling the ones without one. On the first non-holiday weekday after each quarter closes, a scheduled task emails me to request the export while I am at my desk. Scheduled checks later that day pick up the export and run the whole pipeline from a written runbook. I backfilled Q1–Q3 2026 to start, and Q3 alone covered 151 conversations and 2,922 messages. Step-by-step: 1. I started from The Rundown’s AI Audit prompt. 2. I requested my Claude data export and automated the process of finding the export email in Gmail, downloading it through Claude in Chrome, and unzipping it through the desktop app on my machine. 3. I condensed the conversations, split them by month, and used Sonnet subagents to extract structured notes from each month. 4. I used an Opus subagent to synthesize the monthly notes into a quarterly report. Each finding must cite at least three conversations across at least two months, include the strongest counter-evidence, and be labeled OBSERVED or INFERRED. Thin evidence is reported as “insufficient evidence.” 5. I had Claude write the results to a Google Doc with Reports and Progress tabs, including footnoted quarterly reports, cross-quarter comparisons, Stop / Start / Continue / Change recommendations, and a Cross-domain pass. 6. I challenged findings I disagreed with and stored those corrections in the doc and in the versioned prompt, now v1.1. 7. I had Claude use the latest audit and its stored memory about me to rewrite its project instructions, including rules about checking connected sources, maintaining long-running threads in a doc, and naming the number that would settle an open decision. 8. I scheduled the export request for the first non-holiday weekday after each quarter closes, then used scheduled checks and a written runbook to run the pipeline. 9. I backfilled Q1–Q3 2026, with Q3 covering 151 conversations and 2,922 messages. 10. To build your own version, add the evidence rules and OBSERVED/INFERRED labels, split the work into inexpensive monthly extraction and one strong synthesis, write the output to a living doc instead of a chat, and make the final step “rewrite your instructions for me.” That turns the audit from a report into a feedback loop.

I wanted a repeatable way to create images with ChatGPT where we first agree on what the image should be, and only then generate it. So I described my preferred process to Plugin Creator and asked it to turn that into a private plugin called "beeldcreaties" (image creations). It is useful for anyone who wants more control over image generation, especially when small details like a crooked beak or an unusual expression matter. The workflow: 1. I say I want to make an image. 2. ChatGPT asks for a reference image and/or a written prompt. 3. I provide the input. 4. ChatGPT explains what it understood: subject, style, mood, important details, and anything that should stay unchanged. It asks for my approval before generating. 5. I approve the description or refine it. If I refine it, ChatGPT updates its understanding and checks again. 6. Once approved, it generates the image. 7. I give feedback. Any new version goes through the same understanding, approval, and generation loop. That approval step matters to me. A crooked beak, an unusual expression, or an imperfect detail might be exactly what gives a drawing its personality, so I want us to agree on what to preserve before creating a new version. To build your own, select Plugin Creator and try: "Create a private plugin called [name] for this image creation workflow: ask for a reference image and/or prompt, explain what you understood, and wait for my explicit approval before generating. If I refine the idea, update your description and ask again. After generation, accept feedback and repeat the approval process before making another version." The plugin uses ChatGPT's available image generation. What we created is a reusable set of instructions for how we collaborate. My favourite part is that I could describe the interaction I wanted in ordinary language and turn it into something I can call on again.

Drop the task that’s still eating your time, or the thing AI keeps getting wrong for you. Your answers help decide what we teach next 🙌

I built a workflow that turned a single plain-English idea into a live, 24/7 crypto trading bot without being a developer. Claude handled the engineering; I made the decisions. Step-by-step: 1. I gave Claude the goal rather than detailed instructions. My first message was roughly: "Create me a trading bot that runs 24/7, makes faster decisions than a human, takes calculated risks, picks the lowest-fee platform, and has a dashboard with profit graphs. Make it your own." Claude designed the whole system and named it Kestrel. 2. Claude explained the setup options, including exchanges versus on-chain trading, laptop versus cloud hosting, and minimum funds. I chose BNB Chain and PancakeSwap, using a brand-new MetaMask wallet dedicated to the bot so it can never touch my other funds. 3. I let Claude build and test the bot. It wrote the strategy—holding good coins in healthy markets and moving to stablecoins in bear markets—along with hard safety limits, including a daily loss cap, emergency brake, and trailing stops. It also built cheapest-route swap selection and automated tests. When the backtests did not clear the bar we had agreed on, Claude told me honestly instead of curve-fitting. 4. Claude walked me through deploying the bot by renting a cloud server, running it in Docker, and locking the server down with Tailscale and a firewall so nothing was exposed to the internet. I only copied commands and sent screenshots when I got stuck. 5. I went live with a small amount and used a private phone dashboard showing the balance, trades, the bot’s reasoning, and Pause and Sell-everything buttons. 6. I kept improving the system through conversation. I added a “strategy league,” where 10 strategies trade on paper and can be promoted only with my approval, along with self-healing, backups, and one-tap sharing of wins and losses to X. My biggest lesson was to talk to Claude like a capable coworker: give it the goal, make the key calls yourself, and ask it to be honest about the risks.

I build, test, ship, and promote an iPhone and iPad reading app with Claude Code agents, each with a specific job. I wanted a reader for my own PDFs and EPUBs where I could write notes in the margin next to the line they belonged to. Nothing I tried did that well, so I built Helicon, a free reader for iPhone and iPad. I use it every day. I do not write most of the code by hand. I run several Claude Code sessions, each with one responsibility. A development session builds the app in Swift. A marketing session reads the usage numbers and finds places to tell people about the app. A separate designer agent, which did not draw the screens, chooses between two or three prototype variants before any UI change is built. Read-only reviewer agents check finished work against the goals I wrote at the start. The part that made this work is turning every mistake into a check rather than a note. When an agent gets something wrong, we turn the lesson into a test, script, or guard that prevents the same bad step next time. For example, the post script reads the community rules again before every post and stops if they have changed. Today, the app lets me hold the margin so a pin drops beside the line. The pin opens a note that I can type or write with Apple Pencil. Review brings notes back days later, next to their lines. The app also includes dark page themes and a speed-reading guide. Listen, which reads the book aloud with on-device voices, is currently in testing. Step-by-step: 1. I wrote down the goals in plain language: installs, people opening a book, minutes read, and the number of days they return. 2. A development session builds each feature test-first and runs the iOS test suite on a simulator before every push. 3. Before any UI change, the agent draws two or three variants. A designer agent that did not draw them chooses one and explains why. 4. I send each build to TestFlight for a small group of testers. 5. A marketing session reads the PostHog numbers every two hours and tries a new channel whenever a round is flat. 6. I send feedback from posts back to the development session as to-do items. 7. I encode every mistake as a test, script, or guard so it does not happen twice. 8. I keep the irreversible steps for myself: App Store submissions, prices, accounts, and deploys wait for my approval. Helicon on the App Store: https://apps.apple.com/app/apple-store/id6811677202?pt=125973633&ct=rundown&mt=8 Built with help from AI coding tools.

I use AI to turn my weekly Slack updates into a report. This workflow is for anyone who posts updates in Slack and needs a summary at the end of the week. I haven't written out the full walkthrough yet, so these are only the basic actions I currently use. I'll update the post with more detail once I've documented the process properly. Step-by-step: 1. I gather the relevant Slack threads from the week. 2. I use AI to turn those threads into a weekly report. My notes also mention a separate goal: copying my own voice with AI, using it to make music, and releasing that music. I'm still working toward that goal and will share a workflow for it after I've tested it. I haven't measured how much time this saves yet.