Project Explainer for Nontechnical Founders Prompt
Explain an existing project in plain language using actual source files and documented decisions.
What this prompt does
Plain-language architecture
Explain observed components without assuming coding knowledge.
Real user journeys
Trace actions through actual files and integrations.
Documented rationale
Separate recorded decisions from an inferred tradeoff.
What to provide & what you get
- Input
- Relevant project files, the founder’s questions and any documented architecture decisions.
- Expected output
- A FORME.md draft explaining user journeys, system parts, data flow and practical operating questions.
How to use
- Provide relevant files and the product questions you want answered.
- Read the plain-language draft alongside its file references.
- Ask the maintainer to resolve unknown decisions or operating procedures.
Check the result: Each architectural statement should trace to a project file or be clearly labeled as an inference.
Copy. Paste. Run.
v1.0.0-gas1Edit the inputs below, then copy into your AI tool with the requested source material.
Before you start
- Relevant project files, the founder’s questions and any documented architecture decisions.
Worked example · Editorial fixture
Explain the task queue to its founder
Original fictional inputs and a prepared editorial answer. This reusable prompt was not run against the fixture. Use the files and case-specific checks to review your own result.
Example input
Use the same original fictional repository supplied with the onboarding example. Founder question: Where does a new task go, and what would a persistence project change? No decision record explains the stack choice.
Prepared output
FORME.md excerpt A user sends a title to POST /tasks. server.mjs:6-10 reads the body and calls createTask; store.mjs:3 rejects blank titles and lines 4-6 add a trimmed task to an array. GET /tasks reads copies (server.mjs:5 and store.mjs:8). The array is a desk tray: it organizes the current process’s tasks but is emptied when that process ends. The analogy does not imply files are written or shared between server processes. Before adding persistence, decide who owns a task, whether edits need history, and how failed writes are reported. This prototype supplies no accounts or database. These are proposed product questions, not hidden implemented features. Why Node was chosen is unknown: no decision record was supplied. The code shows its current role, not the author’s motive.
What to check in this case
- The founder journey uses the same actual files as the developer guide, but explains data lifetime and product questions in plain language.
- No service prices, deployment history or undocumented choice rationale are added.
Keep in mind
- Reading selected files does not establish complete system coverage.
- The prompt cannot reconstruct undocumented motives, incident history or service pricing.



