20 prompts
Read the app's features, settings and error messages. Write a help centre: getting started, one article per feature, troubleshooting for every error users can hit, and account and billing questions. Plain language, short steps, a screenshot list for each. Save to docs/help/ in a structure ready to publish.
RepoWrite 30 support reply templates for the questions we get most (or are most likely to get, based on the product): login issues, billing, refunds, bugs, feature requests, how-to questions, cancellations. Friendly, short, with placeholders and a link to the right help article. Save to docs/support-macros.md.
RepoDocument our public API from the code: authentication, every endpoint with parameters, example requests in curl and JavaScript and Python, responses, errors, rate limits and pagination. Generate an OpenAPI spec if we do not have one and check it against the code.
RepoRewrite the README: what it is in one line, a screenshot or GIF, quick start in under 5 commands, key features, configuration, how to contribute, and licence. Verify every command works by running it.
RepoDefine the 10 metrics that matter for this business (activation rate, retention by cohort, MRR, churn, conversion by channel and so on) with exact definitions. Write the SQL or analytics queries for each against our schema. Do not run against production; save to docs/metrics.md for me to plug in.
RepoBuild a weekly report template and a script that fills in what it can automatically (commits shipped, issues closed, analytics numbers if an API key is available). Sections: metrics, wins, problems, learnings, next week priorities. Save to ops/weekly-report/.
RepoDraft this month's investor or advisor update from what you can find: shipped features from git, metrics from docs/metrics.md or my notes, and research findings. Sections: highlights, numbers, lowlights, asks. Leave clear placeholders for numbers you cannot verify. Save to ops/investor-update.md.
RepoWrite an onboarding guide for a new developer or contractor: how to set up the project, architecture overview, how we deploy, where secrets live (not the secrets themselves), conventions, and a first-week task list with good first issues. Test the setup steps in a clean checkout.
RepoRead the codebase and git history to identify the 10 most important technical decisions (framework, database, auth, hosting, key libraries). Write a short ADR for each: context, decision, alternatives, and consequences, inferred from the code and commits. Save to docs/adr/.
RepoReview our pricing, costs per user (from infra and API usage in the code), competitor prices in research/pricing.csv, and conversion data if available. Propose 3 pricing changes with expected impact and risk, and how to test each safely. Save to ops/pricing-plan.md.
Repo WebBuild a simple unit economics model: cost to serve one user per month (hosting, AI, email, payments fees from our code and config), price, gross margin, and break-even user counts. Make it a spreadsheet-ready CSV with formulas described. Save to ops/unit-economics.md.
RepoRead our terms of service, privacy policy and cookie policy. Compare them with what the product actually does (data collected, third parties, AI use, payments). List mismatches and missing sections. Draft updated text for review by a lawyer, clearly marked as a draft, not legal advice.
RepoRead all open GitHub issues. Close duplicates and stale ones (with a polite note drafted, not posted), add labels and clear titles, group related issues, and write a prioritised list of the top 15 with effort estimates. Save the plan to ops/issue-triage.md before making any changes.
RepoLook for common questions in our issues, discussions, support macros and community posts. Write a public FAQ page with the 25 best questions and answers, add it to the site with FAQPage structured data, and link it from the footer and pricing page.
Repo WebFind startup programs, cloud credits, AI API credits, grants and accelerator programs we could apply to in the next 3 months (cloud providers, AI labs, regional grants, open source funds). List eligibility, deadline and value. Draft the application answers for the top 3. Save to ops/grants.md.
WebList every service this repo depends on (package.json, env var names, config files, integrations). For each, note what it is used for, whether it looks still used, and a cheaper or free alternative where one exists. Save to ops/tools-audit.md so I can cancel what we do not need.
RepoRead my notes in ops/todo.md (or any TODO or notes files in this repo). Group the tasks, mark which ones Claude Code could do for me without supervision, which need me, and which can be dropped. Then do the first 3 that you can do safely, and report back.
RepoFor each upcoming call I list in ops/calls.md (name, company, purpose), research the person and company from public sources and write a one-page brief: background, recent news, likely priorities, 5 smart questions, and what I want from the call.
WebSet up internationalisation if missing, extract all user-facing strings, and translate them into the 3 languages our traffic or market suggests. Keep strings natural, not literal. Check layouts in the browser with longer translations, and list strings that need a native speaker to review.
Repo BrowserLook at everything in this repo and my notes: open issues, docs/roadmap.md, growth/, research/, ops/todo.md. Write a ranked queue of the 10 best tasks for an unattended overnight Claude Code run tomorrow, each as a complete, self-contained prompt with the files to read, the output to produce, and a clear "done" check. Save to ops/overnight-queue.md.
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.
Synthetic personas, simulated usability tests and interview prep built from your real product. Use them to find obvious problems before you spend time with real users, not instead of them.
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.
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.