Kamy
Other PDF MCP servers ask you to install Chrome, manage a local renderer, and ship templates yourself. Kamy MCP is a hosted endpoint — point any MCP client at https://mcp.kamy.dev/mcp, paste an API key, and start asking your AI to "generate an invoice for Acme Corp". Eight production-ready templates ship with the service (invoice, quote, receipt, contract, agreement, certificate, report, shipping-
README
<p align="center"> <a href="https://kamy.dev"> <picture> <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/rakanalalami/kamy/main/apps/web/public/brand/logo-light.svg"> <img alt="Kamy" src="https://raw.githubusercontent.com/rakanalalami/kamy/main/apps/web/public/brand/logo.svg" width="220"> </picture> </a> </p>
<p align="center"> <a href="https://www.npmjs.com/package/@kamydev/sdk"><img alt="SDK on npm" src="https://img.shields.io/npm/v/@kamydev/sdk?label=%40kamydev%2Fsdk&color=1863dc"></a> <a href="https://www.npmjs.com/package/@kamydev/cli"><img alt="CLI on npm" src="https://img.shields.io/npm/v/@kamydev/cli?label=%40kamydev%2Fcli&color=1863dc"></a> <a href="https://github.com/rakanalalami/kamy/blob/main/LICENSE"><img alt="MIT License" src="https://img.shields.io/npm/l/@kamydev/sdk"></a> </p>
PDF generation API for developers. Ship invoices, receipts, contracts, and reports with gorgeous defaults.
Install
# SDK — in your app
npm i @kamydev/sdk
# CLI — globally for scripts, CI, previews
npm i -g @kamydev/cli
import Kamy from "@kamydev/sdk";
const kamy = new Kamy({ apiKey: process.env.KAMY_API_KEY! });
const pdf = await kamy.render({
template: "invoice",
data: { invoiceNumber: "INV-001", total: 1500, currency: "USD" },
});
console.log(pdf.url); // signed URL, valid 1 hour
What's New
| Date | Change |
|---|---|
| Apr 2026 | GET /v1/renders now returns a per-render cost (in major currency units) plus a top-level currency: "usd". Denormalized at insert time via a Postgres trigger reading users.plan, so the cost reflects your plan when the render happened. Surfaced in SDK 0.2.3 as RenderSummary.cost / RendersListResult.currency. |
| Apr 2026 | SDK now ships OpenAPI-generated types — import type { paths, components } from "@kamydev/sdk/openapi" for fully-typed request/response shapes derived from the live spec. Raw openapi.json shipped too. |
| Apr 2026 | Per-template JSON Schema validation at render time via Ajv (formats enabled). Previously only required was checked — full type/format/enum/oneOf enforcement is now live. |
| Apr 2026 | options.encrypt — AES-256 password protection + permission flags (printing, copying, modifying, annotating, form-fill, accessibility, assembly). |
| Apr 2026 | OpenAPI 3.0.3 spec now covers all 19 /v1/* routes with full request/response schemas — https://kamy.dev/api/openapi.json |
| Apr 2026 | kamy push <file> CLI + client.pushTemplate() SDK + idempotent PUT /v1/templates upsert. Run it on every commit to sync templates from the repo. |
| Apr 2026 | options.validateOnly: true — dry-run any render to validate the payload without consuming quota. |
| Apr 2026 | options.pdfA: "1b" | "2b" | "3b" — best-effort PDF/A archival tagging via embedded sRGB ICC OutputIntent + XMP. |
| Apr 2026 | options.metadata (info dict) + options.watermark (pdf-lib stamp, distinct from theme.watermark). |
| Apr 2026 | options.inlineImages + options.compressImages — server-side image hardening, defends against expiring signed URLs. |
Full per-release detail in CHANGELOG.md.
Architecture
Kamy is a Turborepo monorepo containing:
| Package | Description |
|---|---|
apps/web |
Next.js 15 app — landing page, dashboard, REST API |
apps/mcp |
Cloudflare Worker — MCP server for AI coding agents |
apps/docs |
Nextra docs site |
packages/sdk |
@kamydev/sdk — publishable TypeScript SDK |
packages/templates |
@kamy/templates — PDF template HTML/CSS + Zod schemas |
packages/ui |
@kamy/ui — shared shadcn/Radix UI components |
packages/shared |
@kamy/shared — types, utils, observability, email |
packages/tsconfig |
@kamy/tsconfig — shared TypeScript configs |
Prerequisites
- Node.js 22+
- pnpm 10+
- Supabase CLI (
brew install supabase/tap/supabase) - Wrangler CLI (
pnpm add -g wrangler) — for MCP server
Local Development
# 1. Install dependencies
pnpm install
# 2. Copy env file and fill in values
cp .env.example apps/web/.env.local
# 3. Start local Supabase (requires Docker)
supabase start
# 4. Apply database migrations
supabase db reset
# 5. Generate TypeScript types from schema
supabase gen types typescript --local > packages/shared/src/types/database.ts
# 6. Start all apps in dev mode
pnpm dev
The web app runs at https://kamy.dev
Commands
pnpm dev # Start all apps with Turbopack (hot reload)
pnpm build # Production build all apps
pnpm typecheck # TypeScript check across all packages
pnpm lint # Biome lint + format check
pnpm lint:fix # Biome auto-fix
pnpm test # Run all tests (Vitest unit + Playwright E2E)
pnpm test:unit # Vitest unit tests only
pnpm test:e2e # Playwright E2E tests only
pnpm clean # Remove all build artifacts
Database
Migrations live in supabase/migrations/. They are applied in order by timestamp.
# Reset and reseed local DB
supabase db reset
# Create a new migration
supabase migration new <name>
# Push to remote Supabase project
supabase db push
Deployment
- Web app: Deployed to Vercel on merge to
main - MCP server: Deployed to Cloudflare Workers via
wrangler deploy - Database: Hosted on Supabase Cloud
Environment Variables
See .env.example for all required variables with descriptions.
Tech Stack
- Framework: Next.js 15 (App Router, React Server Components, Turbopack)
- Language: TypeScript 5.7+ strict
- Database: Supabase (Postgres 16 + Auth + Storage)
- Styling: Tailwind CSS v4 (Oxide engine)
- Payments: Stripe (Meters API)
- PDF Engine: Browserless.io v2
- Email: Resend + React Email
- Observability: Sentry + PostHog + Axiom
- Testing: Vitest + Playwright + MSW
<!-- build: 2026-04-22T09:20:52.217Z -->
Open Plugins
This repository ships an Open Plugins distribution so AI coding agents (Cursor, Claude Code, and any other Open-Plugins-compatible tool) can install Kamy as a single plugin and get a ready-to-use PDF pipeline.
The plugin bundles four components at the repo root:
| Path | Component | Purpose |
|---|---|---|
.plugin/plugin.json |
Manifest | Name, version, metadata, declared components |
skills/kamy/SKILL.md |
Skill | Teaches the agent when and how to render PDFs |
rules/kamy-sdk.mdc |
Rule | Coding conventions for @kamydev/sdk and the REST API |
agents/pdf-template-author.md |
Sub-agent | Specialised template-design agent (Handlebars + print CSS) |
.mcp.json |
MCP server | Hosted MCP at mcp.kamy.dev exposing render tools |
To install the plugin in a supported agent, point it at this
repository (https://github.com/rakanalalami/kamy) and follow the
host's plugin-install flow. The agent will read the manifest and
load the components automatically.
See the Open Plugins specification for the full schema.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.