Skip to main content

FREE TOOL

AGENTS.md Generator

Structured agent instructions. Live preview. Clean Markdown export. Shareable links.

Draft the file coding agents read first - build commands, conventions, layout, and security notes. No account. Nothing leaves your browser until you commit the file.

AGENTS.md vs CLAUDE.md vs llms.txt

Live preview

Acme Payments API

TypeScript monorepo for billing, subscriptions, and Stripe webhooks. Prefer small, reviewable PRs with tests for payment and auth paths.

Build commands#

Shell
npm install
npm run build

Test commands#

Shell
npm run test
npm run lint

Code style#

  • Use TypeScript strict mode.
  • Prefer explicit return types on exported functions.
  • Keep modules focused - one domain per folder under src/lib/.

Conventions#

  • Stripe webhook handlers live in src/lib/billing/.
  • Never log secrets or full card numbers.
  • All billing mutations go through subscription-items.ts helpers.

Repository layout#

  • apps/web - Next.js dashboard and API routes
  • packages/shared - shared types
  • supabase/migrations - database schema

PR & commit guidelines#

  • Run the full test suite before opening a PR.
  • Keep PRs to one logical change; billing and auth changes need a test.
  • Title format: [area] Short imperative summary.

Security notes#

  • Do not commit .env files.
  • Webhook routes must verify Stripe signatures before dispatching events.

AGENTS.md is the front door - Moxie keeps what it points to accurate Export free above · connect a repo when you want Moxie to keep it accurate

Agent-first structure

Runnable command blocks, one-rule-per-bullet conventions, repository layout, PR guidelines, and security notes - the structure agents actually parse.

Live preview

See the rendered AGENTS.md as you type. No separate export step to review formatting.

Shareable & portable

The full draft lives in the URL. Send the link or bookmark it. Export clean Markdown ready for your repo root.

Private by design

Everything runs in your browser. Instructions never leave your device until you commit the file.

What to put in AGENTS.md

Build and test commands

The exact commands agents should run before and after edits - install, build, lint, and test.

Conventions and layout

Where modules live, naming patterns, and rules agents should not guess from file names alone.

Security boundaries

Secrets handling, auth paths, and operations that require human review before an agent proceeds.

Pointers to deeper docs

Link to architecture pages or an llms.txt index when agents need more than a single-repo cheat sheet.

Frequently asked questions

Built for the AI-agent era

AGENTS.md is the front door - Moxie keeps what it points to accurate

Draft and export AGENTS.md here. Read how it differs from CLAUDE.md and llms.txt, then connect GitHub when you want Moxie to extract conventions from source and keep linked docs current on every merge.

Export free above · connect a repo when you want Moxie to keep it accurate