You finished the Foundation module with three documents: a product-context.md, a market-landscape.md, and a codebase-map.md for your own repo. This toolkit makes them infrastructure. A context doc you wrote once helps you once; a context doc your team pastes into every chat — that Claude reads on every task — helps every spec, prototype, and memo, forever.
Everything here is copy-able. Replace the Mizan placeholders (the GCC bookkeeping brand whose co-founder, Dana Al-Qassimi, runs through the module) with your own, save the files in the folder your team works in, and open it in Claude Desktop. From then on, every product task in that folder starts from your real user, your real bets, and your real code — instead of each conversation re-explaining the company from scratch.
The whole pack is four files and a handful of prompts. That’s the point — a foundation small enough to actually get pasted beats a beautiful one that lives in a doc nobody opens. Keep the sensitive stuff out of it: this folder holds your strategy and conventions, never the cap table, unreleased financials, or customer PII.
The founders CLAUDE.md
This is the standing context Claude reads automatically when it opens your project folder, so nobody has to re-explain “our user” or “our bets” every chat. Keep it short and current — it points at the foundation docs rather than repeating them.
# CLAUDE.md — [Company name] product & strategy
## What we do
[One-line product summary]. Our north star is [metric]. The full
context lives in product-context.md — read it before specing, deciding,
or planning anything.
## The three source-of-truth files — read these first
- product-context.md → WHAT we're building, WHO it's for, our bets and
our anti-focus
- market-landscape.md → WHERE we win and lose, and the wedge we're
staking the roadmap on (refresh quarterly)
- codebase-map.md → WHERE a feature would actually live in our repo
(re-run when a major module changes shape)
Always check product-context.md before weighing a feature against "our
strategy" — that phrase should point at this file, not a vibe.
## The golden rule
Claude drafts; a named person decides. Every spec, prototype, decision
memo, and board narrative is a draft until a founder or PM accepts it.
An unaccepted draft is a starting point, not a commitment.
## Verification non-negotiables
- A confident gap is better than a confident guess — if the context
doesn't say it, flag it, don't invent it.
- Every claim about our own code names a real file Claude actually read,
not a plausible-sounding guess.
- Every number that reaches a roadmap, a customer, or a board is checked
against a real source before it ships.
## DATA SAFETY — non-negotiable
The cap table, unreleased financials, customer PII, and anything under
NDA stay out of this folder and out of casual chats — keep them in a
workspace the company has approved for that data. This folder holds
strategy and conventions, never the confidential layer.
## What's safe in this folder
- Safe: product-context.md, market-landscape.md (public material only),
codebase-map.md, specs, decision memos, aggregate product metrics.
- Never here: the cap table, unreleased financials, raw customer PII,
anything under NDA, competitor material you agreed not to share. Tell
me if a file looks like it has any of these before we use it.On Desktop you create and edit this in the file pane — no commands. Anthropic’s docs call the “Ask permissions” mode the one recommended for new users; leave it on, and the first time Claude reads your repo or your deck you’ll approve it in the prompt.
The product-context.md template
The one-page premise sheet every later spec, memo, and plan reasons from. The anti-focus section is the discipline — a context doc that’s all positive description hasn’t earned its keep yet.
# product-context.md — [Company name]
## Summary
[One paragraph: what you're building and why now.]
## What we're building & the problem it solves
[Concrete, not aspirational — the specific problem and how you solve it.]
## Who it's for
For: [the one user that matters most right now]
NOT for yet: [who you're explicitly not building for this phase, and why]
## North-star metric
[Metric name]: [current value]. [How you measure it / where it lives.]
## Strategy & bets this phase
- [Bet 1, one line]
- [Bet 2, one line]
- [Bet 3, one line]
## Explicitly NOT doing
- [Cut 1] — [one-line reason]
- [Cut 2] — [one-line reason]
## Hard constraints
Team: [size]. Runway: [months]. Platform: [web/iOS/Android]. Regulatory:
[anything that applies — data residency, payment rails, PDPL, etc.]The market-landscape.md template
The reusable read your roadmap and board narrative both start from. Including the boring non-consumption competitors is what keeps this honest.
# market-landscape.md — [Company name]
## Summary
[One paragraph: the shape of the field.]
## Competitors
[Name] — who they're for: [...]. Where they win: [...]. Where they lose:
[...]. Strong or just loud: [...].
[Repeat, including "do nothing" and "a spreadsheet" where relevant —
the non-consumption baseline is a real competitor for most early
products.]
## Market shifts that matter
- [Shift 1 and why it affects your bets]
- [Shift 2 and why it affects your bets]
## Our wedge
[The position you're staking the roadmap on], because [the real
strength it leans on]. To defend it we have to keep being true that
[what you'd need to back up].
## Watch-list
- [Signal that would change this picture]The codebase-map.md template
The grounded scope you bring into an eng conversation — every line traces back to a file Claude actually opened.
# codebase-map.md — [repo name]
## Shape
[Folder]: [what it's responsible for]. [Repeat for each top-level folder.]
## A real flow, traced
[User action] →
1. [real file] — [what it does]
2. [real file] — [what it does]
3. [where the data ends up]
If [this flow] breaks, start in [file].
## Basic questions answered
- [Question]: [grounded answer, naming the file]
## Candidate feature: [name]
Would likely touch: [real files/components]. Would likely need: [new
files/tables]. Rough size: [small/medium/large] — a starting estimate
for the eng conversation, not a number to defend.The saved-prompt library (Foundation)
The recurring foundation jobs, as prompts you reuse. Paste them in the chat with the right files open; on the Terminal & Automation track these become slash commands so they’re one keystroke, but you need none of that to use them today.
Extract a context doc from what you already have
"Here's our pitch deck and any strategy notes. Reverse-engineer a
product-context profile: what we're building, who it's for, our
north-star metric if you can find it, this phase's bets, and our
hard constraints. Where the source is vague or silent, say so and
ask me — don't invent it."
Map the competitive landscape
"In the market/ folder are competitor pages and my own rough
observations. Map the field for me — direct competitors, adjacent
ones, and the 'do nothing' / 'a spreadsheet' options. For each, one
line on who they're for. Don't rank yet, just lay out the field."
Get the lay of the codebase
"This is our codebase. In plain English, give me a map: the main
folders and what each is responsible for. Keep it to a one-screen
overview — no code yet, just the shape of it."
Trace a real flow
"Walk me through exactly what happens when [a real user action] —
step by step, naming the real files in order. Where would I look
if this were failing?"
Scope a candidate feature against the real code
"If we built [a feature], which existing files would an engineer most
likely touch, and what new files would probably be needed? Just the
rough shape — I'll use this to scope the eng conversation."Installing it (Desktop, no setup)
- Save the four files —
CLAUDE.md,product-context.md,market-landscape.md,codebase-map.md— in the folder your team works in. - Open that folder in Claude Desktop and approve the read in the “Ask permissions” prompt.
- Start any task by opening the relevant docs; the prompts above do the rest.
That’s the whole install. The foundation you built in the module is now shared infrastructure — and Module 2 of the track extends this same pack with the spec-and-prototype engine that builds on it. If you’re standing this up across a team rather than for yourself, read the operating guide for the decision-and-data-safety layer that belongs underneath the whole system.