clawnify
v0.10.0
Published
Deploy your apps and website to Clawnify from the terminal with Claude Code, Codex, and Cursor
Downloads
2,292
Readme
Clawnify CLI
Deploy full-stack software with a database to production in seconds. No infrastructure setup, no config files, no Cloudflare account needed.
Build with your AI coding tool, deploy with one command, share a live URL. Once you sign in to Clawnify, your team can access all their software and agents from one place — with managed auth across everything.
npx clawnify init # app or website (asks)
cd my-project
npx clawnify deployApps go live at https://<slug>.apps.clawnify.com with a D1 database, edge hosting, and a global CDN. Websites go live at https://<slug>.myclawnify.com — one public Astro site per org.
Requirements
- Node.js 22+ — required by the bundled Wrangler toolchain
Quick Start
Create a new app
npx clawnify initPick a template (blank or CRUD), and you get a ready-to-deploy project:
my-app/
src/
server/ # Hono API routes + D1 database
client/ # React frontend (Tailwind v4)
clawnify.json # App manifest
wrangler.toml # Local dev configDevelop locally
cd my-app
pnpm install
pnpm devOpens at http://localhost:5173 with a local D1 database. Same code, same database interface — no conversion between local and production.
Deploy
npx clawnify login
npx clawnify deployThat's it. Your app is live with:
- Edge hosting on Cloudflare Workers
- D1 database (SQLite at the edge)
- Global CDN for static assets
- Custom URL at
*.apps.clawnify.com
If an app with the same slug already exists in your org, the CLI prompts you to update it or create a new one with a suggested slug. After the first deploy, a .clawnify/project.json file is written — subsequent deploys from the same directory update the existing app automatically.
For CI or agent-driven deploys, pass --yes (default to updating on conflict), --update-existing, --create-new, or --org <slug-or-id> to skip the prompt. In a non-TTY environment, an unresolved slug conflict exits with code 75 instead of hanging on input. Use --dry-run to print the file list and gzipped size without uploading.
Install
# Run directly (recommended)
npx clawnify <command>
# Or install globally
npm i -g clawnifyCommands
clawnify init [dir] # Scaffold a new project (asks: app or website)
clawnify init --type website # Skip the prompt, scaffold an Astro site
clawnify login # Authenticate with Clawnify
clawnify deploy [dir] # Deploy to production (app or website, auto-detected from clawnify.json)
clawnify deploy --from owner/repo # Deploy from a GitHub repo (apps only)
clawnify deploy --dry-run # Preview the file list without deploying
clawnify deploy --yes # Non-interactive: update existing app on slug conflict
clawnify deploy --org <slug-or-id> # Deploy to a specific org
clawnify link <app-id-or-slug> # Re-bind .clawnify/project.json (accepts id or slug)
clawnify pull schema --from <id-or-slug> # Fetch deployed schema.sql into the local repo
clawnify org list # List orgs you belong to
clawnify org use <slug-or-id> # Switch the active org (global)
clawnify ls # List your deployed apps
clawnify open <slug> # Open app in browser
clawnify logs <app-id> # View build logs
clawnify rm <app-id> # Delete an app
clawnify whoami # Show current user
clawnify ai-files install # Install rules + skills + MCP into your agent(s)
clawnify ai-files uninstall # Remove themWebsites
clawnify init --type website scaffolds a minimal Astro project:
my-site/
src/
pages/index.astro
layouts/BaseLayout.astro
styles/global.css # Tailwind v4
astro.config.mjs # @astrojs/cloudflare adapter
clawnify.json # { website: { framework: "astro" } }clawnify deploy tars the source, the platform runs astro build, and the site goes live at <slug>.myclawnify.com. Each org has one website — every deploy overwrites it.
Deploy from GitHub
Any repo with a clawnify.json can be deployed:
npx clawnify deploy --from clawnify/open-fieldserviceclawnify.json
Add this to your repo root to make it deployable:
{
"$schema": "https://app.clawnify.com/schema/v1/clawnify.json",
"name": "My App",
"description": "What the app does",
"app": {
"framework": "react+hono",
"database": true
}
}Claude Code / AI agents
Make a project agent-ready in one step:
npx clawnify ai-files installThis writes three things into your project:
- Rules — an always-on Clawnify guidance block in your agent's rules file
(
CLAUDE.md,AGENTS.md,.cursorrules, …) so the agent knows the stack, theschema.tsworkflow, identity headers, and@clawnify/connections. - Skills — the
build-a-clawnify-appskill (and any others the CLI ships) into.claude/skills/,.cursor/rules/, etc. - MCP server — a project
.mcp.jsonwiring the Clawnify MCP server, so Claude Code candeploy,list_apps,get_app, anddelete_appdirectly.
clawnify init runs this automatically after scaffolding (pass --no-ai-files
to skip). It installs into every coding agent detected in the directory, and
defaults to Claude (CLAUDE.md) when none is found. Re-run it after upgrading
the CLI (npm i -g clawnify@latest) to refresh rules and skills in place;
clawnify ai-files uninstall removes everything cleanly.
If you'd rather wire only the MCP server by hand:
claude mcp add clawnify -- npx clawnify --mcpThen Claude Code can deploy apps directly:
> Deploy this app to ClawnifyHow It Works
Your code → clawnify deploy → tar.gz upload → build on Cloudflare
→ D1 database created
→ Schema applied
→ Vite builds frontend
→ Deployed to Workers for Platforms
→ Live at *.apps.clawnify.comLocal dev uses wrangler dev which runs D1 as local SQLite — identical to production. No conversion, no surprises.
Frameworks
| Framework | Stack |
|-----------|-------|
| react+hono | React 19 + Tailwind v4 frontend + Hono API + D1 (default) |
| preact+hono | Preact frontend + Hono API + D1 |
| vite-preact | Vite + Preact SPA |
| static | Static HTML/CSS/JS |
Links
- Website
- Dashboard
- Deploy Button — one-click deploy for any repo with
clawnify.json - JSON Schema — editor autocomplete for
clawnify.json
