What this prompt does
This prompt generates two interconnected documents — a README.md and a CONTRIBUTING.md — from a single structured invocation. It is not a generic "write me a readme" request. The template explicitly sequences the README by user psychology: the first scroll captures attention (logo, badges, one-liner, GIF placeholder), the second justifies the project's existence ("Why this exists"), and everything after converts a curious visitor into an active user or contributor. That ordering is deliberate and mirrors what maintainers at well-adopted OSS projects actually do.
The CONTRIBUTING.md side of this prompt is equally intentional. It covers non-code contributions explicitly (docs, issue triage, discussions), which is the single biggest reason first-time contributors bounce — they assume only code is welcome. The PR template and issue templates are generated inline, not referenced as "coming soon," so you get working scaffolding immediately.
The [alternatives] comparison variable is the honest differentiator here. The prompt asks for a truthful competitive comparison, not marketing copy. That alone builds more trust with technical evaluators than five pages of feature lists.
When to use it
- You are open-sourcing an internal tool and need professional documentation before the first public commit.
- You are reviving a stalled project and want to attract contributors with cleaner onboarding.
- You maintain a library and your existing README was written in "ship it" mode — quick summary, no structure.
- You are starting a new CLI, SDK, or framework and want the contributor infrastructure in place before you announce on Hacker News or Reddit.
- You are a solo dev releasing a side project and want it to look credible without spending a weekend on documentation.
Example output
# Patchwork — Schema Migration Differ for PostgreSQL
> Diff two PostgreSQL schemas and generate safe, reversible migration SQL — in seconds.
<!-- GIF suggestion: terminal screencast of `patchwork diff prod.sql staging.sql` -->
## Why this exists
Manually comparing schema dumps is error-prone and slow.
Existing tools either require a live connection or produce noisy diffs
that mix data changes with structural ones. Patchwork works offline,
focuses only on structure, and outputs SQL you can actually run.
## Quick Start
```bash
npm install -g patchwork-diff
patchwork diff schema_v1.sql schema_v2.sql --out migration.sql
CONTRIBUTING.md opens with: "Thanks for looking at Patchwork. You don't need to write code to help — filing a clear bug report or improving a confusing doc paragraph matters just as much."
## Pro tips
- **Set `[setup_time]` to "5 minutes" or less.** If your actual setup takes longer, fix the setup first. A Quick Start that takes 20 minutes kills adoption before the README lands.
- **Be specific in `[target_contributors]`** — "backend Go devs familiar with database tooling" generates tighter coding standards and commit conventions than "developers."
- **Use the `[project_state]` variable honestly** — "pre-alpha, no tests yet" signals to contributors where effort is actually needed and sets accurate expectations.
- **Run the CONTRIBUTING.md output past a junior dev before publishing.** The prompt generates structure; the terminology still needs a real human check for jargon.
- **Pair this prompt with a GitHub Actions workflow prompt** — the README badges reference CI status but the workflow file is out of scope here. Generate it separately so the badges are live from day one.
