{"slug": "a-stack-that-fits-your-project-and-keeps-fitting-as-it-changes", "title": "A stack that fits your project, and keeps fitting as it changes", "summary": "Manifestack launched a free, MIT-licensed agent skill that installs via `npx skills add manifestack/manifestack` for Claude Code, Cursor, Codex, Copilot, Windsurf, OpenCode and Cline, auditing a project's full stack against budget, user count, data rules and team preferences recorded in a STACK.md file. The tool runs locally with no API keys and read-only access, reading package.json, requirements.txt, go.mod, vercel.json, Dockerfile, infra/*.tf and .env variable names to flag drift such as Supabase Free's 500 MB database cap going read-only, a free-tier Gemini API sending EU user documents, duplicate auth providers (Supabase Auth plus Clerk), and a ~$310/mo Kubernetes setup serving 40 users that a ~$25/mo single host and managed Postgres would cover.", "body_md": "Free agent skill for Claude Code, Cursor and Codex\n\nA stack that fits your project, and keeps fitting as it changes.\n\nFor developers and small teams who pick their own services. It fits the stack to your budget, users, data rules and team, then checks every change: new services, plan limits, bills.\n\nThen ask the agent to run manifestack in your project.\n\n```\nnpx manifestack install --agent codex\n```\n\nThen run $manifestack in your project.\n\n```\nnpx manifestack\n```\n\nDetects your agents and asks where to install: Copilot, Windsurf, OpenCode, Cline and more.\n\n```\nnpx skills add manifestack/manifestack\n```\n\nInstalls the skill without the new-service hook. In Claude Code and Cursor, /manifestack offers to add it.\n\nMIT\n\nNo API keys\n\nRead-only\n\nRuns locally\n\nStacks don't break on day one. They drift out of fit.\n\nA requirement gets missed in a quick setup. The agent adds a service nobody checks. The plan is too small for the load, or the setup far too big. Nobody re-checks the fit until something breaks.\n\nOutgrown\n\nRead-only\n\nSupabase Free caps the database at 500 MB. At 2 MB a day, that's months away. Then a launch, or a log table nobody trims, adds 365 MB in three weeks, and the database goes read-only: no signups, no orders.\n\n500 MB\n\nOn the +2 MB/day trend: week 33Actual: week 11\n\nMissed requirement\n\nEU vs free tier\n\nA prototype sent user documents to the Gemini API on the free tier, and production kept it. Free-tier content can go into improving Google's products, and EU users need the paid tier. Your contract promises no training on customer data.\n\nrequires\n\nno training on customer data\n\nai plan\n\nGemini API, free tier\n\n✕\n\nUnchecked change\n\n2 auth providers\n\nAsked to add login, the agent installs Clerk. Supabase Auth already signs users in. Now users live in two places, and one of them bills per user.\n\n```\n  \"@supabase/supabase-js\": \"^2\"\n+ \"@clerk/nextjs\": \"^6\"\n```\n\nNothing failedNothing was checked\n\nOverbuilt\n\n~$310 vs ~$25\n\nA Kubernetes cluster, a multi-zone database, a NAT gateway and a load balancer, serving 40 users. A single app host and managed Postgres would carry them. Big clouds are the right call at scale. Not at forty users.\n\nBuilt for 1M users~$310/mo\n\nRight-sized for 40~$25/mo\n\nSame app, same 40 users~$3,400/yr extra\n\nThere's no best stack. There's yours.\n\nBudget, users, data rules, what your team prefers and what it won't touch decide the fit, from the framework to the cloud. Manifestack starts there, so the same question gets a different answer.\n\nSolo founder, content site and waitlist\n\nbudget\n\nunder $25/mo\n\nusers\n\n500 → 5k\n\nrequires\n\nnothing special yet\n\nprefer\n\nReact, Firebase\n\navoid\n\nservers to maintain\n\nfits\n\nAstro, static pages, no server to run\n\nSupabase Free, database and auth in one\n\nResend Free\n\nrevisit_when users > 3k\n\nTeam of four, B2B dashboard behind login\n\nbudget\n\n~$600/mo\n\nusers\n\n20k → 100k\n\nrequires\n\nEU database, SOC 2 vendors\n\nprefer\n\nPostgres, AWS\n\navoid\n\nKubernetes, ops on-call\n\nfits\n\nReact + Vite, client-side app, no SSR needed\n\nAPI and Neon Postgres in Frankfurt\n\nResend Pro\n\nrevisit_when monthly bill > $500\n\nIndie mobile app with subscriptions\n\nbudget\n\nunder $100/mo\n\nusers\n\n5k → 50k MAU\n\nrequires\n\nApp Store and Google Play billing\n\nprefer\n\nReact Native, Expo\n\navoid\n\nrunning own servers\n\nfits\n\nExpo with EAS Build and Update\n\nRevenueCat for in-app subscriptions\n\nSupabase Free, database and auth\n\nrevisit_when mau > 40k\n\nIllustrative profiles. Your answers go into STACK.md, and every later audit and every new service is checked against them.\n\nThe whole product, not one package.json.\n\nFrontend, backend, workers, infrastructure and every third-party API. Each service is sorted by role and checked against your requirements and the vendor's pages today.\n\n1 · Reads\n\nweb/package.jsonNext.js · Clerk · Supabase\n\napi/requirements.txtFastAPI · OpenAI · Stripe\n\nworker/go.modStripe · Resend · Sentry\n\nvercel.jsonhosting\n\nDockerfile · infra/*.tfinfrastructure\n\n.envvariable names, never values\n\n2 · Sorts by role\n\nHosting\n\nVercel\n\nDatabase\n\nSupabase\n\nAuth\n\nClerk + Supabase Authsame job twice\n\nPayments\n\nStripe\n\nEmail\n\nResend\n\nMonitoring\n\nSentry\n\nAI\n\nOpenAI\n\n3 · Checks against\n\n.manifestack/STACK.md\n\nbudget, users, requires, prefer, avoid\n\nvendor pages, read today\n\npricing, limits, regions, certifications\n\nFindings\n\nRequirement\n\nOverlap\n\nOverbuilt\n\nLimit\n\nBill\n\nRisk\n\nSet the fit once. Every change gets checked against it.\n\nKnow the project\n\nbudget, users, rules, team\n\nStarts from what the stack has to fit: budget, expected users, data region and certifications, what the team prefers and what it avoids.\n\nCheck the fit\n\n/manifestack audit\n\nReads vendor pages today and reports what no longer fits: requirement gaps, overlaps, overbuilt setups, limits and bills, each with a date, a cost and a fix.\n\nRemember why\n\n.manifestack/STACK.md\n\nKeeps the requirements, decisions and their reasons in your repo, so the next session and the next developer start from the same facts.\n\nCheck every change\n\nhook in Claude Code, Cursor\n\nWhen the agent adds a vendor, it is checked against your requirements, your avoid list and the services you already run, before you rely on it.\n\nEmpty repo\n\nDescribe what you're building: budget, expected users, data rules, what your team prefers and avoids. Manifestack proposes a stack, explains each pick and prices it as you grow.\n\nExample: Next.js on Vercel, Supabase, Resend\n\n1k users\n\n$0/mo\n\n10k users\n\n$65/mo\n\n100k users\n\n~$420/mo\n\n/manifestack init\n\nExisting project\n\nPoint it at any repo, yours or a client's: frontend, backend, workers and infrastructure. It maps what's wired up and checks it against your requirements and usage: data region, duplicated services, overbuilt setups, limits and costs.\n\n/manifestack audit\n\nEvery change\n\nSay you ask the agent to add login and it installs Clerk. The hook checks the new SDK against STACK.md, and the agent tells you before it finishes:\n\n+ \"@clerk/nextjs\": \"^6\"\n\nClerk (auth) added. Supabase Auth already signs users in: an Overlap. Clerk bills per user; check it against the ~$600/mo budget.\n\nhook in Claude Code, Cursor\n\nEvery finding says when it matters and how to fix it.\n\nOutput of /manifestack audit on a sample product: a Next.js frontend, a Python API and a Go worker, 9,000 monthly users, seven services. Requirement gaps, duplicated services, launch risks, limits and bills, each with a fix you can act on.\n\n~/relaylog/manifestack auditSample output\n\nFindingWhenCostFix\n\nRequirement\n\nSupabase project region vs. EU database rule\n\nProject in us-east-1, STACK.md requires EU\n\nWhen\n\nBefore launch\n\nCost\n\n~1 day of migration\n\nFix\n\nCreate the production project in an EU region and move the data while it is still small.\n\nOverlap\n\nTwo auth providers: Clerk and Supabase Auth\n\n@clerk/nextjs and supabase.auth both in web/src/\n\nWhen\n\nNow\n\nCost\n\n$25/mo on Clerk Pro, more past 50,000 MRU\n\nFix\n\nBoth do the same job here. Keep the one whose features you use and remove the other from the code.\n\nRisk\n\nOpenAI usage tier 1: tokens per minute\n\napi/ summarizes every upload; the launch-day estimate is above the tier limit\n\nWhen\n\nLaunch day\n\nCost\n\nHigher tiers need paid history and time\n\nFix\n\nRequest a tier increase now, queue uploads in the worker, and use a smaller model for short summaries.\n\nLimit\n\nSupabase Free: database size\n\n312 / 500 MB, +1.1 MB/day\n\nWhen\n\nETA ~Mar 2027\n\nCost\n\n$25/mo on Pro\n\nFix\n\nArchive events rows older than 90 days, or schedule the upgrade for February.\n\nBill\n\nVercel Pro: data transfer\n\n3.4 TB last month, 1 TB included\n\nWhen\n\nEvery month, growing\n\nCost\n\n~$360/mo overage\n\nFix\n\nCache /og/* images at the edge and set Cache-Control on /api/feed.\n\nFix first\n\nEU region, before launch\n\nSpend to review\n\n$385/mo ($4,620/yr)\n\nChecked against\n\nEU, SOC 2 · 7 pages, Oct 6\n\nUsage figures come from numbers you give it, exports in the repo or a read-only vendor MCP. A report is as current as those inputs.\n\nManifestack commands\n\nTwo to start. Plain /manifestack picks the right one: init for an empty repo, audit for an existing one. In Codex, use $manifestack; in other agents, ask for the skill by name.\n\nCommand\n\nWhat it does\n\nExample\n\nAvailable now\n\n/manifestack init\n\nNew project: asks what the stack has to fit, proposes one priced at 1k, 10k and 100k users, and writes the first STACK.md.\n\n/manifestack init \"B2B dashboard, EU users\"\n\n/manifestack audit\n\nExisting project: checks every service against your requirements and growth and reports what no longer fits, each with a when, a cost and a fix.\n\n/manifestack audit\n\nComing next\n\n/manifestack compareplanned\n\nTwo vendors side by side for your workload: features, limits, region, certifications, lock-in, price.\n\n/manifestack compare neon supabase\n\n/manifestack migrateplanned\n\nA step-by-step plan for a vendor or plan switch, with a rollback.\n\n/manifestack migrate supabase --to pro\n\n/manifestack watchplanned\n\nRe-reads vendor pages for the services in STACK.md and proposes a diff when something changed.\n\n/manifestack watch\n\n.manifestack/STACK.md\n\n```\n## Requirements\n\nbudget: ~$600/mo\nusers: 9k now, 50k by Q3\nrequires: EU database, SOC 2 vendors\nprefer: Postgres, Next.js\navoid: Kubernetes  # no ops on-call\n\n## Database: Supabase\n\nplan: free\nlimit: 500 MB database, pauses after 7 days idle\nsource: supabase.com/pricing  # read 2026-10-06\nusage: 312 MB, +1.1 MB/day (2026-10-06)\ndecided: stay on Free until first paying user\nrevisit_when: db_size > 400 MB OR date >= 2027-02-01\nnext: Pro, $25/mo\nenv: SUPABASE_URL, SUPABASE_ANON_KEY  # names only\n\n## Email: Resend\n\nplan: pro, $20/mo\nlimit: 50,000 emails/month\nrevisit_when: monthly_sent > 45,000 OR before launch\n```\n\nOne file, versioned with your code.\n\nSTACK.md lives in .manifestack/, the only folder the skill writes to. The skill tells your agent to read it before changing infrastructure. You review it in pull requests.\n\nrevisit_when\n\nA condition, not a reminder. Every audit checks it and flags the section once it's true.\n\nStored\n\nYour fit: budget, users, data rules, what the team prefers and avoids\n\nServices, plans and the limits that apply\n\nUsage snapshots with the date they were taken\n\nDecisions and the reason behind each one\n\nEnv var names, so the agent knows what is wired\n\nNever stored\n\nSecret values, API keys or tokens\n\nConnection strings or passwords\n\nCustomer data or anything from your database\n\nBilling details from vendor accounts\n\nYour keys never leave your machine. Neither does your code.\n\nBy default it needs no account access, and its only network calls read vendors' public pages: pricing, limits, regions and certifications. A read-only vendor MCP is your own connection, used only if you set it up.\n\nYour machine\n\nRepo scan and analysisruns in your agent session\n\n.manifestack/STACK.mdwritten to your repo\n\nUsage numbersyou give them, or a read-only vendor MCP\n\nNever sent anywhere by Manifestack\n\nSource code\n\nAPI keys and .env values\n\nAccount logins or OAuth access\n\nUsage and billing data\n\nGET public pagespublic page only\n\nVendor public pages\n\nPricing, regions and security pages, such as supabase.com/pricing. Read during the audit like any visitor would. Requests carry no project data.\n\nYour agent's model provider sees what your agent already sees. Manifestack adds no new destination.\n\nPricing, limits and regions, read from the source.\n\nPage maps for hosting, database, auth, email, storage, payments, mobile, monitoring and AI. Each audit reads the vendor's current pages for plans, limits, regions and certifications. No rankings, no affiliate links, no \"switch to X\".\n\nThe skill finds the vendor's public pages during the audit. A page map makes it faster; new ones ship as releases.Request a map\n\nFrequently asked questions\n\nWho is it for?\n\nDevelopers, small teams and agencies who build with Claude Code, Cursor or Codex and choose their own services: hosting, database, auth, email, payments, AI APIs. It helps most from the first commit to a few hundred thousand users, when nobody owns infrastructure full-time.\n\nHow is this different from asking my agent?\n\nA plain agent answers from training data that can be a year old, then forgets the answer. Manifestack starts from your project's requirements, reads vendors' current pages, keeps every decision in STACK.md, and checks each later change against them.\n\nWill it change my code, config or vendor settings?\n\nNo. The audit is read-only. The only thing it writes is .manifestack/STACK.md, and decisions in it (plan, revisit dates) only after your yes. Installing a package or changing a setting happens only when you ask your agent for it.\n\nDo I need to give it API keys or account access?\n\nNo. It reads your repo and env var names, never their values, and asks you for usage numbers. If you've connected a vendor's official MCP in read-only mode, such as Supabase, Neon or PostHog, it can read usage there instead. Manifestack itself never asks for a key or a login.\n\nDoes it work with Cursor, Codex or my agent?\n\nYes, if your agent reads Agent Skills (SKILL.md). Cursor, Codex, GitHub Copilot, Windsurf and OpenCode do, and npx manifestack finds them for you. The new-service check is dependable in Claude Code and Cursor, where a hook flags every vendor SDK the agent adds. In other agents it depends on the agent picking up the check on its own, so run /manifestack audit before you merge a change that adds or swaps a service.\n\nDoes it work with Python, Go or mobile apps?\n\nYes. It reads dependencies from package.json, requirements.txt, pyproject.toml, Pipfile and go.mod, plus config files and env var names in any project. React Native and Expo apps are covered, including RevenueCat, Adapty and EAS. Native iOS and Android dependencies (Podfile, Gradle) are not read; the skill asks about services it cannot see.\n\nWhere do prices and limits come from?\n\nFrom each vendor's public pages (pricing, limits, regions, certifications), read during the audit. Every finding cites the page and the date it was read. If a page can't be fetched, the finding says so instead of guessing. Needs an agent with web access.\n\nWhat does it check besides costs?\n\nThe fit itself: data region and certifications against your requirements, services that do the same job, setups far bigger than the load, and what the team prefers or avoids. init proposes a whole stack and explains each pick, including when Next.js or a big cloud is more than you need. Compliance decisions stay yours.\n\nMy vendor isn't on the list. Is it still useful?\n\nYes. The skill looks up its public pages and records the service, plan and limits in STACK.md. A page map makes it faster and more reliable. To add one, open an issue with the pricing page.\n\nIs it free?\n\nYes. MIT license, no paid tier, no account. There is no Manifestack server to pay for.", "url": "https://wpnews.pro/news/a-stack-that-fits-your-project-and-keeps-fitting-as-it-changes", "canonical_source": "https://manifestack.com/", "published_at": "2026-10-07 12:13:33+00:00", "updated_at": "2026-10-07 12:21:12.205709+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "ai-tools"], "entities": ["Manifestack", "Claude Code", "Cursor", "Codex", "GitHub Copilot", "Windsurf", "OpenCode", "Cline"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/a-stack-that-fits-your-project-and-keeps-fitting-as-it-changes", "markdown": "https://wpnews.pro/news/a-stack-that-fits-your-project-and-keeps-fitting-as-it-changes.md", "text": "https://wpnews.pro/news/a-stack-that-fits-your-project-and-keeps-fitting-as-it-changes.txt", "jsonld": "https://wpnews.pro/news/a-stack-that-fits-your-project-and-keeps-fitting-as-it-changes.jsonld"}}