DEV Community: Mikhail Bogovalov

RendrKit: The Open-Source Alternative to Bannerbear

Mikhail Bogovalov — Tue, 17 Mar 2026 08:30:58 +0000

If you've ever needed to generate images programmatically - social media cards, OG images, blog headers, infographics - you've probably come across Bannerbear.

Bannerbear is solid. You design a template, call their API, get an image back. They're doing around $40-50K MRR, plans start at $49/mo, and they've earned it. Placid and HTMLCSStoImage do similar things in slightly different ways.

But I kept running into the same friction: I didn't want to design templates from scratch every time. I didn't want to pay $49/mo just to experiment. And most importantly - I was building AI agents that needed to generate images, and none of these tools were built with that in mind.

So I built RendrKit.

What RendrKit does differently

RendrKit is a Design API for AI Agents. Not just image generation - AI-native image generation.

Here's what that means in practice:

80+ templates built-in - no need to design anything before your first API call
MCP server - AI agents can discover and use templates directly
Direct render mode - skip the queue, get images at $0.001/img
Free tier - 50 images/mo, no credit card

The core idea: your AI agent should be able to say "I need a social media image with this title and background" and just... get one. Without you hand-crafting templates first.

Show me the code

Basic REST API call

curl -X POST https://api.rendrkit.dev/api/v1/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "templateId": "social-post-bold",
    "slots": {
      "title": "RendrKit just launched",
      "subtitle": "Generate images with a single API call",
      "background_color": "#1a1a2e"
    }
  }'

Direct render with template + slots

const response = await fetch("https://api.rendrkit.dev/api/v1/generate", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    templateId: "blog-header-minimal",
    slots: {
      title: "How We Cut Image Generation Costs by 98%",
      author: "Jane Doe",
      avatar_url: "https://example.com/avatar.jpg",
      tag: "Engineering",
    },
    render: "direct", // skip the queue, get the image immediately
  }),
});

const { url } = await response.json();
console.log(url); // your generated image

That's it. No template builder, no drag-and-drop editor, no 15-minute setup. Pick a template from the gallery, fill the slots, get an image.

The AI agent angle

This is where it gets interesting. RendrKit ships an MCP server, which means any AI agent that speaks MCP can:

Browse available templates
Understand what slots each template expects
Generate images autonomously

If you're building with Claude, GPT, or any MCP-compatible agent - your agent can generate branded visuals without human intervention. That's the "Design API for AI Agents" positioning - it's not just an image endpoint, it's a creative tool that AI can use natively.

SDKs and ecosystem

npm: npm install rendrkit - docs
PyPI: pip install rendrkit - docs
MCP server: plug into any MCP-compatible AI agent
REST API: works with anything that can make HTTP requests

Pricing comparison

Here's an honest side-by-side:

	Bannerbear	Placid	RendrKit
Free tier	No	Limited	50 images/mo
Starter	$49/mo	$29/mo	Free
Pro	$149/mo	$99/mo	$49/mo
Enterprise	$199/mo	Custom	$149/mo
Built-in templates	No (build your own)	Some	80+
MCP server	No	No	Yes
Direct render	No	No	$0.001/img

I'm not going to pretend RendrKit does everything Bannerbear does. Bannerbear has years of polish, a mature template editor, and great docs. If you need a full visual template builder with pixel-perfect control, Bannerbear is excellent.

But if you want to hit an API with a template ID and some text and get an image back - especially from an AI agent - RendrKit is built for that.

Use cases

What people are building with RendrKit:

Social media images - automated posting pipelines that generate visuals per post
OG images - dynamic Open Graph images for blog posts and pages
Blog headers - consistent branded headers without opening Figma
Infographics - data-driven visuals generated from dashboards
E-commerce - product cards, sale banners, promo images at scale

Try it

Free tier, 50 images/mo, no credit card: rendrkit.dev

Docs: rendrkit.dev/docs

Browse templates: rendrkit.dev/templates

If you have questions or feedback, I'm around - drop a comment or find me on the site. I'm genuinely curious what people build with this.

Build an Auto-Posting Instagram Bot in 20 Lines of Python

Mikhail Bogovalov — Sun, 01 Mar 2026 12:59:15 +0000

What if your AI agent could generate professional Instagram posts and schedule them automatically — with zero design skills?

I built RendrKit, a Design API that turns text into production-ready graphics. Today I'll show you how to combine it with Python to create an auto-posting Instagram bot in ~20 lines of code.

What We're Building

A Python script that:

Takes a topic (e.g., "productivity tips")
Generates a professional Instagram graphic via API
Posts it to Instagram automatically
Runs on a schedule (daily, hourly, whatever)

Here's the end result — generated entirely from text:

Prerequisites

pip install rendrkit instagrapi requests

A free RendrKit API key (50 images/month on free tier)
An Instagram account

The Code (Yes, It's Actually 20 Lines)

from rendrkit import RendrKit
from instagrapi import Client
from datetime import datetime

# Setup
rk = RendrKit(api_key="your-rendrkit-key")
ig = Client()
ig.login("your_instagram", "your_password")

# Generate image from a text prompt
result = rk.generate(
    prompt="A motivational quote about productivity: 'Ship fast, learn faster.' Modern dark gradient design with bold typography",
    size="1080x1080"
)

# Download the image
import requests
image_path = "/tmp/post.png"
with open(image_path, "wb") as f:
    f.write(requests.get(result["url"]).content)

# Post to Instagram
ig.photo_upload(image_path, "Ship fast, learn faster 🚀\n\n#productivity #startup #buildinpublic")
print(f"Posted at {datetime.now()}")

That's it. Run it, and you've got a fresh Instagram post generated from nothing but text.

Level Up: Use Templates for Consistent Branding

The prompt mode is great for one-offs, but for a branded feed you want consistency. RendrKit has 69+ templates — pick one and fill in the slots:

result = rk.generate(
    template_id="social-quote",
    slots={
        "heading": "Ship fast, learn faster.",
        "subheading": "The best feedback comes from real users.",
        "accent_color": "#6366F1"
    },
    size="1080x1080"
)

Same API, but now every post matches your brand colors and layout. No more random AI-generated chaos.

Make It Automatic: Run Daily with Schedule

import schedule
import time
import random

tips = [
    {"heading": "Ship fast, learn faster.", "subheading": "Real users give the best feedback."},
    {"heading": "Done > Perfect", "subheading": "Perfection is the enemy of progress."},
    {"heading": "Build in public.", "subheading": "Your journey inspires others."},
    {"heading": "Solve your own problem.", "subheading": "The best startups come from personal pain."},
    {"heading": "Talk to users daily.", "subheading": "Metrics lie. Conversations don't."},
]

def post_daily():
    tip = random.choice(tips)
    result = rk.generate(
        template_id="social-quote",
        slots={**tip, "accent_color": "#6366F1"},
        size="1080x1080"
    )
    with open("/tmp/post.png", "wb") as f:
        f.write(requests.get(result["url"]).content)
    ig.photo_upload("/tmp/post.png", f"{tip['heading']}\n\n#startup #buildinpublic #productivity")
    print(f"Posted: {tip['heading']}")

schedule.every().day.at("09:00").do(post_daily)

while True:
    schedule.run_pending()
    time.sleep(60)

Set it and forget it. Your Instagram posts itself every morning at 9 AM.

Why Not Just Use DALL-E or Midjourney?

Good question. I tried that first. Here's why templates win for social media:

	DALL-E / Midjourney	RendrKit Templates
Consistency	Every image looks different	Same brand, every time
Text rendering	Misspells words constantly	Pixel-perfect typography
Speed	5-15 seconds	1-2 seconds
Cost	$0.04-0.08/image	$0.001/image (direct render)
Brand control	Hope for the best	Exact colors, fonts, layout

For generative art? Use DALL-E. For professional branded content? Use templates.

Other Ideas You Can Build

This same pattern works for:

Twitter/X auto-poster — generate OG images for your threads
Email newsletter headers — fresh header image every issue
Slack bot — /motivate command that generates and posts a quote
E-commerce — auto-generate product banners from your catalog
Certificates — generate completion certificates for your course platform

Available Templates

RendrKit has 69+ templates including:

social-quote — motivational quotes, tips
blog-post-cover — article headers
product-showcase — product announcements
stats-banner — metrics and numbers
hiring-post — job announcements
event-poster — event promotions
certificate — completion certificates
invoice — business documents

Full list: rendrkit.dev/templates

Get Started

Grab a free API key at rendrkit.dev
pip install rendrkit
Start generating

The free tier gives you 50 images/month — enough to post daily and test everything.

I'm building RendrKit as a solo dev. If you have questions or feature requests, hit me up on Twitter/X or drop a comment below.

Add Image Generation to Your AI Agent in 5 Minutes

Mikhail Bogovalov — Fri, 27 Feb 2026 13:21:48 +0000

Your AI agent can write emails, summarize docs, and answer questions. But ask it to create a social media image? 🦗

That's what I built RendrKit for - a Design API that lets any AI agent generate professional images with one API call.

The Problem

AI agents are great at text. Terrible at visuals. If your agent needs to:

Create an Instagram post for a client
Generate an OG image for a blog
Make a banner for a product launch

...you're stuck calling Canva manually. That defeats the purpose of automation.

The Solution: One API Call


bash
curl -X POST https://api.rendrkit.dev/api/v1/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Instagram post about morning coffee routine, warm aesthetic",
    "size": "1080x1080"
  }'

Response: a CDN URL with your image. Done.

Two Modes

Prompt mode - describe what you want in plain English. GPT picks the best template and fills it.

Direct render mode - you pick the template and slots yourself. $0.001 per image. Full control.

{
  "templateId": "bold-title-overlay",
  "size": "1200x628",
  "slots": {
    "title": "Ship Fast, Break Nothing",
    "subtitle": "A developer's guide to velocity",
    "accent_color": "#6C5CE7"
  }
}

MCP Server (for Claude, Cursor, etc.)

npx @rendrkit/mcp@latest

Add to your Claude Desktop config:
{
  "mcpServers": {
    "rendrkit": {
      "command": "npx",
      "args": ["-y", "@rendrkit/mcp@latest"],
      "env": {
        "RENDRKIT_API_KEY": "your_key"
      }
    }
  }
}

Now Claude can generate images directly in conversation.

What You Get

• 50+ professionally designed templates
• 5 sizes (1080x1080, 1200x628, 1080x1920, 1200x1200, 1280x720)
• Automatic Unsplash/Pexels background photos
• Logo support
• REST API + MCP server + npm SDK
Pricing

• Free: 50 images/month
• Starter: $19/month - 500 images
• Pro: $49/month - 2,000 images
• Business: $149/month - 10,000 images
Try It

1. Sign up at rendrkit.dev
2. Get your API key
3. Make your first image in 30 seconds

Feedback welcome 🙏

How I Built a Design API for AI Agents as a Solo Dev

Mikhail Bogovalov — Wed, 25 Feb 2026 11:53:42 +0000

AI agents are getting powerful at text - but they still can't make images. No social posts, no banners, no infographics. I built RendrKit to fix that.

What It Does

One API call with a text prompt returns a production-ready PNG:


bash
curl -X POST https://api.rendrkit.dev/api/v1/generate \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Launch announcement for RendrKit - Design API for AI agents"}'

Behind the scenes:

1. GPT-4o-mini analyzes the prompt and picks the best template
2. It fills in the template slots (headline, subtitle, colors, photo query)
3. Playwright renders the HTML/CSS to PNG
4. Image goes to CDN, you get a URL back
Two Modes

Prompt Mode - send text, get image. GPT does the thinking. $0.005/image.

Direct Render - you pick the templateId and fill slots yourself. No GPT, no latency. $0.001/image.

The Stack

• Next.js on Vercel for API routes
• Playwright for HTML-to-PNG rendering
• GPT-4o-mini for template selection + slot filling
• Supabase for auth, usage tracking, API keys
• Unsplash/Pexels for background photos
50+ Templates

Social posts, banners, cards, infographics, event posters, testimonials, real estate listings - all responsive across 5 sizes (1080x1080, 1200x628, 1080x1920, 1200x1200, 1280x720).

Integrations

• REST API
• npm SDK (@rendrkit/sdk)
• MCP server (@rendrkit/mcp)
• OpenAI function calling schema
What I Learned

• Playwright on serverless is tricky but doable (Vercel has a Chromium layer)
• GPT-4o-mini is good enough for template selection at $0.005/image margin
• HTML/CSS templates beat AI image generation for structured content - predictable, fast, brandable
Check it out: rendrkit.dev

Free tier: 50 images/month. Would love feedback!