Spec Driven Development
by formulahendry
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Runs on your computer★ 440MITUpdated 14 months ago
Install
Foundation
Overview
Description from the maker (original English)
Feasibility studies with computed costs, break-even, NPV and IRR.
PopNx MCP’s three pillars
These three are never merged into one score: opinions, measurements and popularity are different things.
- User voiceiNo reviews yetReviews open together with accounts in a later step
- Our basic checksi1 of 1 passedChecked from public information; code not scanned yet
- UsageiNot measured yet★ –
What you can do with it
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
- Connection: Runs in the cloud
- Transport: Streamable HTTP
- Category: Other
Install
Pick the AI app you use, then copy the settings.
Before you install
- Basic checks: 1 of 1 passed
- Runs on the maker’s server: what you ask it goes through their server.
- The registry lists no key to fill in.
You need
- Runs in the cloudNothing to install; it connects over the internet
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Works in Claude on the web, desktop and mobile:
- Open Settings → Connectors
- Press Add custom connector
- Paste the address below and press Add
- If it asks you to sign in, sign in with the maker
https://getfndtn.com/api/mcpRun this in your terminal
claude mcp add --transport http foundation https://getfndtn.com/api/mcpOr copy the settings yourself
Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"foundation": {
"url": "https://getfndtn.com/api/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Or copy the settings yourself
Paste into .vscode/mcp.json in your project
{
"servers": {
"foundation": {
"type": "http",
"url": "https://getfndtn.com/api/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
ChatGPT connects only to servers on the internet:
- Open ChatGPT on the web → Settings → Connectors (Apps)
- Press Create or Add (some plans need Developer mode under Advanced first)
- Paste the address below, give it a name and save
- If it asks you to sign in, sign in with the maker
https://getfndtn.com/api/mcpMenu names can differ slightly by plan and over time.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
What it can do
We have not read this server’s tool list yet; we check endpoints in rotation.
Basic checksi
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 7, 2026
- ✓ Pass
Maker matches the repo
The registry name was verified with the GitHub account mamarexa, which owns the repo
- … Not checked yet
Still maintained
We have not read this repo from GitHub yet
- … Not checked yet
Has a licence
We have not read this repo from GitHub yet
- … Not checked yet
Endpoint responds
Not checked yet; we check endpoints in rotation
- … Not checked yet
Sign-in with OAuth
Not checked yet; we check endpoints in rotation
- … Not checked yet
No hidden instructions in tool descriptions
Not checked yet; we read the tool list when we check the endpoint
Compared with others in Otheri
- More stars than 0% of the other 32655 servers in this category
- Published to the registry more recently than 54% of the other 32655 servers in this category
- Basic checks equal to or better than 100% of the other 32655 servers in this category
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
Common questions
Is Foundation safe?
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
What keys or codes does it need?
The registry does not list a required key, though some servers ask you to sign in on first use.
Does it run on my computer or in the cloud?
It runs in the maker’s cloud. Connect over the internet with nothing to install, but what you send passes through the maker’s servers.
Is it still maintained?
The GitHub repo last changed –, and the latest version (1.0.0) reached the registry 34 days ago.
Which apps can use it?
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.
What can I use instead?
In the same category: Spec Driven Development, Remote Desktop Commander, SuperCollider
Alternatives
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
Remote Desktop Commander
by desktop-commander
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Runs in the cloud★ 211Updated 2 days ago
Install
SuperCollider
by Tok
Make sound with SuperCollider: synths, drums, chords, melodies and soundscapes over OSC.
Runs on your computerUpdated today
Install
README from the maker
Open the README
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Foundation Skills
Open-source agent skills that do a feasibility consultant's job: size the market, benchmark competitors, build the financial study, tune the pricing, then write the business plan and the pitch deck. They run in your own AI agent (Claude, Codex, Gemini CLI, Cursor and others). They are built by the team behind Foundation.
▶ Foundation in 48 seconds: a sample café project from its costs to its business plan, and an AI assistant testing it.
Why this exists
Ask a general AI assistant whether your business idea works and it will write you something that sounds like a feasibility study. The problems are underneath:
- The math is guessed. Language models are not calculators. NPV, IRR, payback and break-even figures in an AI-written plan are often wrong, and nothing on the page tells you which ones.
- The assumptions are invisible. Where did "£38,000 for the fit-out" come from? A lender will ask, and "the AI said so" is not an answer.
- The parts don't agree. The team section says five staff, the payroll line pays four, and the market chapter uses a different number from the pitch deck.
A professional consultant avoids all three. They put a source behind every input, run the numbers through a real financial model, and keep one version of the truth that every document draws from. These skills teach an AI agent to work the same way.
How it works
The skills split the job the way it should be split:
- The agent does the analyst's work: interviewing the founder, researching costs, prices and competitors, choosing a sensible structure, tagging where every number came from, and writing.
- A deterministic engine does the math: Foundation is a feasibility-study web app with a tested calculation engine. It computes investment, operating costs, revenue ramps, profit, cash flow, NPV, IRR, payback, break-even and SaaS metrics from the inputs. When your agent is connected to Foundation, every financial figure in the output comes from that engine. The agent never computes one itself.
The skills also work without a Foundation account. The agent still does the research and structures every input with its source, and never invents financial results. It then hands you a foundation-project.json file. Import that file into Foundation (Projects → Import) to get the computed study.
The skills
| Skill | What it does |
|---|---|
feasibility-analyst | Turns a description of your business into a complete feasibility study: business type, cost categories, capital and operating costs, staffing, revenue model and financial assumptions, each input with its source. Then it reads the results and gives you a go / no-go memo with the risks and levers. |
market-research | Sizes the market bottom-up (TAM, SAM, SOM) with a top-down cross-check. Profiles competitors from public sources and fills a competitors × features benchmark. Checks whether your revenue plan needs more of the market than you can realistically win. When a price isn't public, it prepares an outreach kit for you to make the enquiries yourself. |
pricing-optimizer | Finds the price range between your cost floor, competitor prices and the value to the customer. Designs three to five distinct pricing options, tests each through the engine with an explicit assumption about sales volume, and recommends one with a plan to validate it. |
business-plan-writer | Writes a lender- and investor-grade business plan as a Word file (.docx). Financial tables, charts and scenarios are generated from the engine's output, so no figure is retyped. |
pitch-deck-writer | Builds the pitch deck as a PowerPoint file (.pptx): one idea per slide, the takeaway as the headline, engine-computed financial slides, and a visual check of every slide before delivery. |
All five follow the same rules, written down in CONVENTIONS.md:
- Every input carries a source tag: founder, quote, benchmark or estimate.
- The agent asks you only what it can't research, in one round.
- Writing follows a guide based on Wikipedia's Signs of AI writing: plain sentences, no hype, no em dashes.
- For brand templates or heavier design, the Word and PowerPoint files can go through your environment's own document tools as a second step, with a script that checks no number changed on the way.
Install
Any agent that supports skills (Claude Code, Codex, Gemini CLI, Cursor and others):
npx skills add mamarexa/foundation-skillsOr copy the skill folders you want by hand:
| Agent | Where skills go |
|---|---|
| Claude Code | ~/.claude/skills/ (all projects) or .claude/skills/ (one project) |
| Claude.ai / Claude Desktop | Zip a skill folder, then Settings → Capabilities → Skills → Upload |
| Codex | ~/.codex/skills/ |
| Gemini CLI | ~/.gemini/skills/ or .gemini/skills/ |
| Cursor | ~/.cursor/skills/ or .cursor/skills/ |
The Word and PowerPoint skills need Python with python-docx, python-pptx and matplotlib. These come preinstalled in Claude's code sandbox. Anywhere else, run pip install python-docx python-pptx matplotlib.
Connect Foundation (optional, recommended)
- Create a free account at getfndtn.com.
- Add
https://getfndtn.com/api/mcpto your agent as a remote MCP server (in Claude: Settings → Connectors → Add custom connector). You'll sign in with your Foundation account. - Ask your agent for a feasibility study. It will find the Foundation tools and work on your real project, and the numbers will match what you see in the app.
Step-by-step setup for Claude, ChatGPT, Codex, Cursor, Gemini CLI, Antigravity and VS Code is in the Foundation docs.
Try it
"I want to open a specialty coffee shop near Leeds station, about 30 seats, opening next
spring. Is it viable, and how much do I need to raise?"
"Size the market for scheduling software for UK dental practices, and benchmark the main
competitors."
"Turn my Foundation project into a business plan for a bank loan, then a 12-slide deck."
Contributing
This repository is meant to get better with use. The most valuable contributions are:
- Benchmarks with sources: cost ratios, startup-cost ranges, wages, employer payroll costs, rents and tax rates for your industry or your country. Local numbers help founders outside the US and UK most.
- Corrections from practitioners: if you do this work for a living and a step is wrong or missing, tell us how it's done in practice.
- Test cases: a realistic business description and what a good study of it should contain.
Good contributions also flow back into the Foundation app, for example as better default categories and benchmark ranges. See CONTRIBUTING.md for the rules. The short version: every number needs a source you'd be comfortable showing a lender.
This repository and the skills/ folder in the Foundation app are kept in sync automatically in both directions, so a merged pull request here shows up there (as a reviewed PR, not a silent merge) without anyone copying files by hand.
Tests
pip install python-docx python-pptx matplotlibpython tests/test_scripts.py
The fixture project in tests/fixtures/ is a sample café. Its facts file is real output from Foundation's engine.
License
MIT. By contributing, you agree your contribution is licensed the same way, including its use in Foundation.