20 prompts
Read this repo and the product docs to understand what we sell and who buys it. Brainstorm 40 link magnets (free tools, calculators, generators, checkers, templates, datasets, interactive explainers) our audience would search for and bookmark. For each, give the long-tail keyword it targets, why people would link to it, how it leads back to the product, and build effort (S/M/L). Cut to the best 20 and save them ranked to growth/link-magnets.md with the top 3 marked "build first".
WebTake the top idea from growth/link-magnets.md (or pick the strongest small free tool for our audience if that file does not exist) and build it as a new page in this codebase. It must work without sign-up, load fast, have a clear H1 that matches the keyword, 300+ words of genuinely useful explanation below the tool, an FAQ with 5 real questions, a soft CTA to the product, and correct title, meta description, canonical and Open Graph tags. Follow existing code patterns, then run the build and lint.
RepoList every number our users calculate by hand or in a spreadsheet because of the problem we solve (costs, ROI, conversion rates, time saved, sizing, pricing). Pick the 10 with real search demand. For each, write the formula, the inputs and sensible defaults, a worked example, and the target keyword. Then build the 3 best as calculator pages using the existing components, each with an explanation section and FAQ.
Web RepoFind a pattern where we could make 50 to 500 useful pages from structured data, such as "[tool] alternatives", "[integration] + our product", "[job title] templates" or "[city] [service] costs". Propose 3 patterns with the keyword shape, the data source, what makes each page unique enough to not be thin content, and a sample of 10 URLs. Pick the best one, build the data file and a page template, and generate the first 20 pages with internal links between them and a sitemap entry.
Web RepoWork out which documents, checklists, spreadsheets, prompts or config files our audience copies from blog posts every week. Write 15 genuinely good templates (not filler), each with a one-line use case, when to use it, the template itself in copy-paste form, and a filled-in example. Save them to content/templates/ as one markdown file each, plus an index page draft that targets "[topic] templates".
Design a small original data study we could publish that journalists and bloggers would cite. Use public data (APIs, open datasets, GitHub, public pricing pages, our own anonymised aggregate numbers if the repo has them) and actually collect it. Produce the dataset as CSV, 5 headline findings with numbers, 3 charts described or rendered, a methodology section, and a 1,200-word write-up draft in content/study/. Flag any finding that rests on a small sample.
WebBuild a glossary of the 60 terms people in our niche search for definitions of. For each term write a clear 2-sentence definition, a 150-word explanation with an example, related terms, and a natural link to the product where it fits. Group terms by theme, and output a data file the site can render as /glossary and /glossary/[term] pages. Skip any term where we have nothing better to say than Wikipedia.
WebResearch the 15 to 25 main tools in our category from their public sites and docs. Build an honest, sortable comparison table: pricing tiers, free plan limits, key features, integrations, platforms, best for. Cite the source URL and date checked for every cell. Save the data as JSON and draft a page "Best [category] tools compared (updated [month year])" with a short verdict for each. Be fair to competitors, including where they beat us.
WebDesign a free "grader" or "checker" for our niche: the user pastes a URL, text, file or config and gets a score with specific fixes (think website graders, headline analyzers, config linters). Define 10 to 15 scoring rules with clear pass/fail logic and the fix text for each. Then build it as a page in this repo that runs client-side where possible, shows the score and fixes, and ends with a CTA. Add tests for the scoring rules.
RepoPick the single concept in our space that beginners find hardest to understand. Build an interactive explainer page: a step-by-step visual with controls the reader can change, plain-language text alongside each step, and a summary at the end. Use the frameworks already in this repo, keep it accessible (keyboard, contrast, reduced motion), and target the "how does [concept] work" keyword.
RepoDesign a free generator (names, policies, bios, briefs, configs, prompts, schemas) that solves a small job our audience does often. The output must be good enough to use as is, and shareable by URL so people send it to colleagues. Build it in this repo with a permalink for each result, a "made with" footer link back to us, and metadata so shared links preview well.
RepoCollect 80 to 100 recent, sourced statistics about our industry and audience (published within the last 2 years where possible). Verify each by opening the original source, not a roundup that quotes it. Group them by theme, write a short intro for each group, and save as content/stats.md with the source link and year next to every stat. Writers link to stats pages constantly, so accuracy is the whole point: drop anything you cannot trace to a primary source.
WebCreate a curated "awesome list" for our niche: 100+ resources (tools, libraries, guides, communities, datasets, people) grouped into clear sections with a one-line honest description each. Check every link works. Save it as a README.md ready for a public GitHub repo, with contribution guidelines, and suggest 10 places to announce it. Include our product only once, in the section where it genuinely fits.
WebDesign something other sites can embed that links back to us: a badge, a live stat, a mini calculator, a status widget or a "powered by" card. Build the embed (a script or iframe plus the snippet users paste), a page that generates the snippet, and make sure the embed is lightweight, works on any site, and includes a followed link back. Explain who would embed it and why.
RepoMine People Also Ask, Reddit, Stack Overflow, Quora and forums for 50 specific questions our audience asks about the problem we solve. Cluster them, and for the 15 clusters with the clearest intent write a draft answer page: a direct answer in the first 2 sentences, a detailed explanation, an example, and a related-questions section. Save to content/answers/.
WebOutline a free 7-lesson course that teaches our audience the skill our product makes easier. Write all 7 lessons in full (800 to 1,200 words each, with exercises), the landing page copy targeting "[skill] course" and "learn [skill]", and the 7 matching emails. Save everything to content/course/. The product should appear only where it truly saves a step.
Find something in our niche that can be measured and ranked (speed, price, accuracy, popularity, uptime) and is not already well covered. Design a repeatable benchmark: what is measured, how, how often, and how ties and bias are handled. Run the first round with real data, save results as JSON, and draft a leaderboard page plus a script that can refresh it on a schedule.
Web RepoWrite the definitive one-page cheat sheet for the main tool, language, framework or workflow our audience uses. It must fit on one printed page, be grouped logically, and include the commands or shortcuts people actually forget. Produce it as markdown and as a print-friendly HTML page with a download button, targeting "[tool] cheat sheet".
Go through every public page in this codebase. Find pages that could become link magnets with modest work: a blog post that could become a tool, a docs page that could become a reference, a feature page that could become a free version. List the top 10 with what to change, the keyword it would target, and effort. Save to growth/upgrade-existing-pages.md.
RepoList the dates, events, reports, launches and trends in our industry over the next 12 months that cause search spikes. For each, propose a page we could publish 4 to 6 weeks earlier (a tracker, guide, template, or prediction piece), with the keyword and a publish-by date. Save as a content calendar to growth/seasonal.md and fully draft the next 2 due.
Web
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
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.
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.