20 prompts
Read the codebase, docs and marketing pages to list every problem, feature, integration and use case we cover. Turn them into a keyword map of 150 keywords grouped into topic clusters, with search intent (informational, commercial, transactional), a rough difficulty guess with your reasoning, and which existing page should rank for each (or "new page needed"). Save to seo/keyword-map.csv and a summary of the 10 biggest gaps.
Web RepoAudit this codebase for technical SEO: titles and meta descriptions (missing, duplicate, too long), canonicals, robots and noindex, sitemap coverage, broken internal links, heading structure, image alt text, structured data, Open Graph tags, hreflang if relevant, and pages that render content only client-side. Write findings to seo/technical-audit.md sorted by impact, then fix the safe ones in a branch and list what you changed.
RepoPick the 10 highest-value missing pages from seo/keyword-map.csv (or research them if it does not exist). For each, write a brief: target keyword and variants, search intent, what the top 5 results cover and miss, a suggested H1 and outline, questions to answer, internal links to add, and a unique angle only we can offer. Save each to seo/briefs/[slug].md.
WebTake 5 briefs from seo/briefs/ and write each article in full. Lead with the answer, use concrete examples, include original screenshots or code where it helps (describe what to capture if you cannot), avoid filler intros and generic conclusions, and link to the product only where it saves the reader real effort. Each article 1,500 to 2,500 words. Save drafts to content/drafts/ and list the facts I should double-check.
Find every blog post or guide in this repo older than 12 months. For each, check whether facts, screenshots, prices, versions or links are now out of date (verify against current sources). Rewrite the 5 most valuable ones with updated information, a new "updated [month year]" note, better structure and stronger answers to the main question. Keep the URLs the same.
Web RepoMap every content page in this codebase and the topics it covers. Find pages that mention a topic we have a dedicated page for but do not link to it, orphan pages with no internal links in, and hub pages that should link to their cluster. Add contextual internal links with descriptive anchor text (no "click here"), at most 5 new links per page. Output a table of every link added.
RepoReview every page type on this site and add the right schema.org JSON-LD: Organization, WebSite with SearchAction, BreadcrumbList, Article, FAQPage, Product, SoftwareApplication, HowTo or ItemList where the content genuinely matches. Use real values from the page, not placeholders. Validate the output shape, and write a short note on which rich results each page type could now get.
RepoList every indexable page with its current title and meta description. Rewrite each to match the main search intent, include the primary keyword naturally, stay under 60 and 155 characters, and give a reason to click (a number, a year, a benefit, a free tool). Apply the changes in code and output a before/after table so I can review.
RepoFor each of our 5 biggest competitors, research their product, pricing, strengths and common complaints (G2, Capterra, Reddit, Twitter/X, app store reviews). Draft a "[Competitor] alternatives" page that lists 6 to 8 real alternatives, including us, with honest pros and cons, who each is best for, and pricing. Save to content/alternatives/. Never invent claims about competitors; cite sources.
WebWrite a "[Our product] vs [Competitor]" page for each of our top 5 competitors: a 2-sentence verdict, a feature table, pricing compared, where they win, where we win, and who should pick which. Pull facts from their current public pages and cite them. Keep the tone fair, because biased comparison pages do not rank or convert. Save to content/vs/.
WebPick the 3 topics most tied to buying our product. For each, design a cluster: one pillar page and 10 to 15 supporting pages, with the keyword, intent and angle of each, and how they interlink. Mark which pages already exist in this repo. Save to seo/clusters.md and write the pillar page for the most important cluster in full.
Web RepoFor our 10 most important pages, search the main keyword and look at what Google currently ranks: format (list, tool, guide, product page), length, angle and freshness. Tell me for each page whether our format matches the intent, and if not, exactly what to change. Save to seo/intent-check.md.
WebFind 20 queries in our niche where a featured snippet or AI overview shows. For each, note the current format (paragraph, list, table) and the page holding it. Write a better snippet-shaped answer we could place on one of our pages (40 to 60 words for paragraphs, clean steps for lists), and say which page it belongs on.
WebResearch how our product and category appear in AI answers (ChatGPT, Perplexity, Claude, Google AI Overviews) for the 20 questions buyers ask. Note which sources get cited. Then recommend concrete changes so we get cited more: pages to create, facts to state plainly, comparison tables, an llms.txt file, third-party sites to get mentioned on. Write the llms.txt and the top 3 page changes.
Web RepoFind every image in this codebase. Add or fix descriptive alt text, rename unhelpful filenames where safe, make sure images have width and height set, use modern formats and lazy-load below the fold. List images over 200KB with a suggested fix. Output a table of changes.
RepoRun Lighthouse (or analyse the code if you cannot run it) on the 5 most important pages. Find the biggest causes of slow LCP, layout shift and slow interaction: render-blocking scripts, huge bundles, unoptimised images, fonts, third-party tags. Fix what is safe in a branch, measure again, and report the before/after numbers.
Repo BrowserIf our product serves specific industries, roles, locations or use cases, list the 20 segments with the most search demand. For each, write a landing page draft that uses that segment's language, problems and examples (not a find-and-replace of the same page). Save to content/segments/ with the keyword each targets.
WebSearch Reddit, Hacker News, Indie Hackers, Stack Exchange and niche forums for threads from the last year where people ask about the problem we solve. Collect 100 real questions in the asker's own words, with the link and upvote count. Cluster them, and for each cluster suggest a page, a doc update or a product change. Save to seo/questions.csv and seo/questions-summary.md.
WebCrawl the blog, docs and resource pages of our biggest competitor (respect robots.txt). List every topic they cover that we do not, and estimate which ones bring them traffic based on how prominent and linked they are. Give me the 25 best gaps with a suggested angle that beats their version. Save to seo/content-gap.md.
WebDevelopers and power users search for "how to [do X] with [our product]" and integration guides. List 20 such tasks from our features and support questions. Write 5 of them as complete how-to docs pages with steps, code or screenshots to capture, common errors and fixes. Save them where the docs live in this repo.
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.
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.