@uge/payo 0.1.7 → 0.3.0

package/README.md

@@ -2,7 +2,7 @@
 
 <img src="https://raw.githubusercontent.com/uttam-gelot/payo/main/assets/logo.png" alt="Payo" width="200" />
 
-**Generate project-tailored AI assistant rules & skills in under two minutes.**
+**Generate project-tailored AI assistant rules & skills in minutes.**
 
 Payo interviews you about your stack, then writes the guidance files your AI
 coding assistant reads — so Claude, Cursor, Copilot, and friends follow _your_
@@ -24,6 +24,7 @@ project's conventions instead of guessing.
 ## Contents
 
 - [What is Payo?](#what-is-payo)
+- [Why Payo?](#why-payo)
 - [Who is this for?](#who-is-this-for)
 - [Quick Start](#quick-start)
 - [How to use — a walkthrough](#how-to-use--a-walkthrough)
@@ -32,6 +33,7 @@ project's conventions instead of guessing.
 - [AI vs. template generation](#ai-vs-template-generation)
 - [Bootstrap prompt](#bootstrap-prompt)
 - [Resume anytime](#resume-anytime)
+- [Roadmap](#roadmap)
 - [Requirements](#requirements)
 - [Run locally](#run-locally)
 - [Contributing](#contributing)
@@ -49,6 +51,38 @@ Where the assistant ships a headless CLI, Payo drives that tool's own AI to
 write rich, project-specific docs; where it doesn't, Payo falls back to solid
 templates — so you always end up with usable output.
 
+## Why Payo?
+
+Every AI coding session starts cold. Your assistant doesn't know you use Drizzle
+over Prisma, that handlers live in `src/routes`, that you write Vitest specs
+beside the file, or that commits follow Conventional Commits. So you re-explain
+it — in chat after chat — and still get code that ignores half of it.
+
+The fix is the guidance files each tool already reads (`CLAUDE.md`,
+`.cursorrules`, `.github/copilot-instructions.md`, `AGENTS.md`). But writing them
+by hand is tedious, easy to get wrong, and different for every tool. Most people
+never do it, or do it once and let it rot.
+
+Payo writes them for you in minutes — tailored to your actual stack,
+in each tool's native format — so your assistant follows _your_ conventions from
+the first prompt instead of guessing.
+
+### Vibe coding, without the mess
+
+Vibe coding is fast and fun — you prompt, the AI improvises, code appears. The
+problem isn't the speed, it's that the assistant has no rules to improvise
+_within_. So it guesses: a different ORM than the rest of the repo, files
+scattered wherever, its own naming, your tests and commit conventions skipped.
+Fine for a throwaway demo; on a real codebase it quietly piles up inconsistency
+you pay for later in reviews, bugs, and refactors — and each new chat starts the
+guessing over.
+
+Payo gives that improvisation a guardrail. It generates the guidance files your
+assistant already reads, so the AI reaches for _your_ framework, _your_ folder
+layout, _your_ testing and git rules every time — without you stopping to explain
+them. You keep vibe coding at full speed; the output just fits the project
+instead of fighting it.
+
 ## Who is this for?
 
 - **Devs starting a new repo** who want their AI assistant productive from
@@ -150,6 +184,14 @@ A few environment variables tune AI generation:
 | `PAYO_RETRIES`          | `1`      | Extra attempts after a failed run    |
 | `PAYO_AGENT_TIMEOUT_MS` | `120000` | Wall-clock cap per file (ms)         |
 
+### Your data stays yours
+
+Payo has no backend and no telemetry. Rich generation runs entirely through the
+AI CLI **you** already have installed (`claude`, `cursor-agent`, etc.), so your
+answers go only to that tool under your own account and credentials — exactly as
+if you'd prompted it yourself. Payo itself sends nothing to any server; without a
+CLI present it never leaves your machine at all, falling back to local templates.
+
 ## Bootstrap prompt
 
 An empty repo with great guidance is still an empty repo. After generating, Payo
@@ -164,6 +206,27 @@ Interrupt a run and pick up where you left off. Payo saves your questionnaire
 answers and generation progress under `.payo/`; rerun and it resumes, only
 generating what's missing. Finished runs clean the directory up automatically.
 
+## Roadmap
+
+Payo works today, but it's still early. Here's where it's headed:
+
+- **First-class existing-project support.** Right now Payo shines on a fresh
+  repo. The next big step is making it just as good on an established codebase:
+  **detect** the stack from what's already there (manifests, lockfiles, config,
+  folder layout), **auto-answer** the questionnaire from that evidence, and let
+  you confirm or tweak instead of typing it all out — then the normal generation
+  flow continues.
+- **Broader stack coverage.** More languages, frameworks, ORMs, databases, and
+  AI tools, plus deeper, more opinionated defaults for the ones already
+  supported — so the guidance fits more of the ecosystem out of the box.
+- **A smoother flow.** Faster, clearer prompts; better defaults and grouping;
+  tighter summaries and confirmations; and a generation step that's quicker and
+  more transparent about what it's doing.
+
+Have a stack you want supported or an idea to make the flow better?
+[Open an issue](https://github.com/uttam-gelot/payo/issues) or see
+[Contributing](#contributing).
+
 ## Requirements
 
 **To run Payo:** [Node.js](https://nodejs.org) **>= 18**. That's it — `npx @uge/payo`