# ShotAPI — Web Screenshot & Render MCP Server for AI Agents Give your AI Agent eyes. One command, zero install. [![GitHub stars](https://img.shields.io/github/stars/smallhandsome/shotapi-mcp-server)](https://github.com/smallhandsome/shotapi-mcp-server) [![MCP Registry](https://img.shields.io/badge/MCP-Registry-blue)](https://registry.modelcontextprotocol.io/servers/io.github.smallhandsome/shotapi-mcp-server) [中文文档](README_CN.md) ## Quick Start (Remote Mode — No Install) ```bash # Claude Code claude mcp add --transport streamable-http shotapi https://aiphotoshop.mynatapp.cc/mcp # Or with npx npx -y @anthropic-ai/claude-code@latest --mcp https://aiphotoshop.mynatapp.cc/mcp ``` For Claude Desktop / Cursor, add to your config: ```json { "mcpServers": { "shotapi": { "type": "streamable-http", "url": "https://aiphotoshop.mynatapp.cc/mcp" } } } ``` ## Real Use Cases ### 1. Claude verifies its own code ``` Human: Build a landing page with a hero section, testimonials, and pricing grid. Claude: [writes HTML/CSS code] Claude: Let me check this with the render tool... [Claude calls ShotAPI render tool with the HTML it wrote] Claude: The pricing grid overlaps on mobile. Fixing responsive breakpoints... ``` ### 2. Deployment verification ``` Human: I just deployed. Check if it looks right. Claude: [calls ShotAPI screenshot on https://yourapp.com] Claude: Footer text is cut off — CSS overflow issue. Background image 404. ``` ### 3. UI regression in CI/CD ```bash # Capture before/after screenshots and compare curl -s "https://aiphotoshop.mynatapp.cc/v1/screenshot?url=https://staging.myapp.com" -o before.png curl -s "https://aiphotoshop.mynatapp.cc/v1/screenshot?url=https://staging.myapp.com" -o after.png ``` ## 3 MCP Tools | Tool | Description | Example | |------|-------------|---------| | `screenshot_one_liner` | One URL, one screenshot | `screenshot_one_liner(url="https://github.com")` | | `screenshot` | Full control: viewport, full-page, CSS selector, format | `screenshot(url="https://github.com", selector="#hero")` | | `render` | Render HTML/CSS to image — Agent writes code, sees result instantly | `render(html="

Hello

")` | The `render` tool closes the feedback loop: **write code → render → see → revise → render again**. ## Why ShotAPI? - **MCP-native** — No glue code. Agent calls tools directly in conversation. - **Zero install remote mode** — streamable-http, no Python/Playwright needed. - **Direct China access** — No VPN required. - **Free tier** — 50 screenshots + 50 renders/month, IP-based, no signup, no API key. All features unlocked (full-page, selector, WebP). - **Render capability** — Not just screenshots of existing pages. Render Agent-generated HTML. - **Built-in ad blocking** — Cleaner screenshots, less noise. - **Listed in MCP Registry** — Official Anthropic MCP server directory. ## Try It Now (No Signup) ```bash curl -s "https://aiphotoshop.mynatapp.cc/v1/screenshot?url=https://github.com" -o shot.jpg ``` ## Pricing | Plan | Price | Limit | |------|-------|-------| | Free | $0 | 50+50/month (IP-based, all features, no signup) | | Starter | $4.90/mo | 5,000/month | | Pro | $9.90/mo | 20,000/month | Free tier: no API key needed. Paid: get key at [pricing page](https://aiphotoshop.mynatapp.cc/pricing). ## Local Mode (STDIO) For private deployments or paid-tier usage, run locally: ```bash # Install dependencies pip install -r requirements.txt playwright install chromium # Set env vars export SHOTAPI_BASE_URL=https://aiphotoshop.mynatapp.cc export SHOTAPI_KEY=your_key_here # optional for paid tiers # Add to Claude Code claude mcp add shotapi python mcp_stdio.py ``` ## Links - **Docs**: https://aiphotoshop.mynatapp.cc/docs - **Pricing**: https://aiphotoshop.mynatapp.cc/pricing - **Health**: https://aiphotoshop.mynatapp.cc/health - **MCP Registry**: https://registry.modelcontextprotocol.io/servers/io.github.smallhandsome/shotapi-mcp-server - **Smithery**: https://smithery.ai/server/@ljs/shotapi - **awesome-mcp-servers**: https://github.com/punkpeye/awesome-mcp-servers - **Dev.to Article**: https://dev.to/smallhandsome/give-your-ai-agent-eyes-building-a-visual-mcp-server-for-web-screenshots-3f49