20 prompts
Create 5 distinct personas who match our target users (different skill levels, goals, devices, patience). For each, walk through our main flow in the running app in the browser, acting as that persona, and narrate what they expect, notice, misunderstand and give up on. Take screenshots at each confusing moment. Write a report ranking the 10 biggest usability problems with evidence and a suggested fix. Save to research/usability-sim.md. Note clearly that this is simulated and should be checked with real users.
Browser RepoBuild 8 realistic personas from what we know about our customers (sales/icp.md, reviews, forum posts). Run a simulated 30-minute discovery interview with each, using good interview technique: past behaviour, not hypotheticals. Write each transcript, then synthesise the patterns, surprising points and the hypotheses we should test with real people. Save to research/synthetic-interviews/.
WebAct as someone who has never heard of us, arriving from a Google search. Open the homepage in the browser, and in under 2 minutes of reading decide: what is this, who is it for, what does it cost, and what do I do next. Write down every moment of doubt. Then go through sign-up and onboarding as that person. Save the walkthrough with screenshots to research/first-time.md.
BrowserFrom the product, reviews of us and competitors, and forum discussions, write a jobs-to-be-done map: the main job, related jobs, the steps of the job, the outcomes users measure success by, and where current tools fail. Mark where our product helps and where it does not. Save to research/jtbd.md.
Web RepoWrite a 30-minute customer interview script for each of our 3 personas: warm-up, past behaviour questions, the switching story, pains, and wrap-up. Avoid leading questions and anything that asks people to predict the future. Add a note-taking template and a screener survey to recruit the right people. Save to research/interview-kit.md.
Design a 10-question survey for our users that tells us why they signed up, what they use most, what nearly stopped them, and what they would pay for. Include answer options, logic, and how each answer would change a decision. Then write the code to analyse a CSV export of responses. Save to research/survey/.
RepoCollect every public review and comment about us and our top 3 competitors (G2, Capterra, app stores, Reddit, X, Product Hunt comments). Pull out the exact phrases people use for their problems, desired outcomes and frustrations. Group them and save the best 100 quotes with sources to research/voice-of-customer.md. These are for copywriting, so keep the original wording.
WebTest our main flows as 4 users with different needs: a screen reader user, a keyboard-only user, someone with low vision using 200% zoom, and someone with a slow connection on an old phone. Use the browser and accessibility tooling (axe, keyboard tabbing, zoom, network throttling). Report what blocks each person and the fix. Save to research/accessibility-sim.md.
Browser RepoCreate 10 buyer personas at different company sizes and budgets. For each, look at our pricing page and simulate their reaction: which plan, whether they would buy, what would make them hesitate, and the price at which it feels too expensive. Summarise into a Van Westendorp-style range and 3 pricing experiments to run for real. Save to research/pricing-sim.md.
RepoRead the onboarding code and any analytics events in this repo. Map every step from sign-up to first value, count the fields, clicks and waits, and list where users are most likely to drop off and why. Propose 10 changes ranked by expected impact and effort, and the events we would need to measure each. Save to research/onboarding.md.
RepoFor the homepage, pricing page and main feature page, take a screenshot and simulate a 5-second test with 5 different personas: what they remember, what they think the product does, and whether they would keep reading. Compare with what we want them to take away. Save findings and copy fixes to research/five-second.md.
BrowserList every page and feature in our app and site. Simulate an open card sort with 6 personas who group them in ways that make sense to them, and name the groups. Compare with our current navigation and propose a clearer information architecture. Save to research/card-sort.md.
RepoWrite 3 to 4 personas for our product backed by evidence: each claim linked to a review, forum post, interview note or analytics finding. Include goals, frustrations, tools used, buying role, and a day in their life. Mark anything that is a guess. Save to research/personas.md.
WebCollect every feature request we can find: GitHub issues, support notes in the repo, TODO comments, public roadmap, community posts. Deduplicate, cluster by underlying need, count requests, and estimate effort from the code. Give me a ranked list with a recommendation for each (build, decline, workaround). Save to research/feature-requests.md.
Repo WebLook at the words we use in the UI (buttons, menus, settings, errors) and compare them with the words users use in reviews and forums. List every term that is jargon, internal naming or ambiguous, with a plain alternative. Apply the clear wins in code and list the rest for me. Save to research/terminology.md.
Repo WebFind every error message, empty state and failure path in this codebase. For each, simulate the user who hits it: what were they trying to do, do they know what went wrong, do they know how to fix it? Rewrite the messages that fail, and list error states that have no message at all. Apply the copy changes.
RepoRun the same simulated usability test on our top competitor's public sign-up and onboarding flow in the browser (do not create accounts that break their terms; stop at walls). Compare step counts, time to value, clarity and polish against ours, with screenshots. Save to research/competitor-ux-test.md.
BrowserOpen our site and app in the browser at iPhone SE, iPhone 15 and a mid-range Android viewport. Go through the main flows as a mobile user on the go. List layout breaks, tiny tap targets, horizontal scroll, slow pages and anything that only works with a mouse, with screenshots. Fix the CSS issues that are safe to fix.
Browser RepoWrite the queries (do not run against production) to find churned users and their last actions. Then, using reviews, cancellation flow copy, and support patterns, list the top 10 likely churn reasons and simulate an exit interview with a persona for each. Propose fixes and a cancellation survey. Save to research/churn.md.
RepoGather every research artefact in this repo (notes, personas, interviews, survey results, usability reports) into a single research/README.md: what we learned, the evidence, confidence level, open questions, and what we changed because of it. Flag findings that came from simulation so they are not mistaken for real-user evidence.
Repo
How to run these while you sleep
- Open your project and start a fresh branch or git worktree, so the whole night is one diff you can review or delete.
- Add your product context above and keep overnight mode on. Claude will log its assumptions instead of waiting for answers.
- Paste one prompt into Claude Code, or save a few to a
queue/folder and loop over them headless. - In the morning, read the files it wrote and the diff. Merge what is good and send what needs sending yourself.
Want it off your laptop? Claude Code on the web runs each session in a cloud sandbox, so you can start a few before bed and close the lid.
Queue a night of prompts
git switch -c overnight/$(date +%F)
mkdir -p logs
for f in queue/*.md; do
claude -p "$(cat "$f")" \
--permission-mode acceptEdits \
--allowedTools "WebSearch" "WebFetch" "Bash(npm run *)" \
> "logs/$(basename "$f" .md).log"
doneOnly allow the tools a task needs. Browser prompts also need a browser MCP server such as Playwright; find one in the MCP directory.
More spare compute prompts
Free tools, calculators, templates and data pages that rank for long-tail searches and earn links on their own. Run these overnight and wake up to shippable pages.
Keyword research, content briefs, technical SEO audits and article drafts you can leave running overnight. Each prompt saves its output to a file so nothing gets lost.
Find link opportunities, draft pitches and build lists of podcasts, newsletters and journalists. These prompts do the slow research so you only do the sending.
Get listed everywhere your buyers look: startup directories, app stores, marketplaces, launch sites and review platforms. Claude prepares every listing so submitting takes minutes.
Lead lists, buying signals and the places your next customers are already asking for help. Claude does the digging; you decide who to contact.
Cold emails, follow-ups, sales pages and objection handling written from what your product actually does. Claude drafts; you review and send.
A month of posts, threads, carousels and replies drafted from your product, changelog and lessons learned. Wake up to a content queue instead of a blank page.
Demo videos, short-form scripts, animated explainers and screenshots generated with code. Use Remotion, Playwright and ffmpeg to render while you sleep.
Teardowns, pricing tracking, feature matrices and screenshots of competitor user flows. Leave a browser running overnight and come back to a full picture of the market.
Rewrite copy, audit the sign-up funnel, design A/B tests and build new landing pages. These prompts work directly in your codebase so the output is ready to review as a pull request.
Welcome sequences, onboarding emails, newsletters and re-engagement campaigns written from your actual product, plus the templates to send them.
Polish passes, onboarding fixes, feature specs and prototypes. Good for overnight runs because each one ends in a branch or a document you can review over coffee.
Tests, refactors, type safety, dependency upgrades and dead code removal. The classic overnight job: long, careful, and easy to review in the morning as a pull request.
Security reviews, secrets scans, load tests, bundle diets and backup checks. The work everyone puts off, done while your machine would otherwise sit idle.
Documentation, help centre articles, support macros, metrics, finance tidy-ups and the admin that piles up. Hand it to Claude when you have compute to spare.
Make Claude better at these jobs with skills, MCP servers and mods, or turn a prompt you run often into a skill with the skill generator.