Get started
Ask your AI (MCP)
Connect your AI tool once, then ask for screenshots in plain words. No code. It works through MCP, the standard way AI tools reach outside services.
What you need
- An API key. Create one in the dashboard, under Connect.
- An AI tool that can add a remote MCP server over HTTP.
Almost every tool connects the same way: the address of our server, and your key in the Authorization header.
A few assistants have no place to put a key. They connect by sign-in instead: the tool opens a localscreenshot window, you sign in and allow it. The table below says which way each tool uses.
Connect your tool
Pick yours. Each family has its own page with the exact line or steps to copy.
| Tool | Family | Connects with | Status |
|---|---|---|---|
| ChatGPT | Assistants | Sign-in (OAuth) | From its docs |
| Claude | Assistants | API key | From its docs |
| Claude Code | Coding tools | API key | Tested |
| Cursor | Coding tools | API key | From its docs |
| Codex | Coding tools | API key | Tested |
| Gemini CLI | Coding tools | API key | Connected |
| VS Code | Coding tools | API key | From its docs |
| Windsurf | Coding tools | API key | From its docs |
| Cline | Coding tools | API key | From its docs |
| Zed | Coding tools | API key | From its docs |
| Copilot CLI | Coding tools | API key | From its docs |
| Hermes Agent | Coding tools | API key | Connected |
| n8n | Automation | API key | From its docs |
| Zapier | Automation | API key | From its docs |
| LangChain | Frameworks | API key | Tested |
| Le Chat | Assistants | API key | From its docs |
| Copilot Studio | Assistants | API key | From its docs |
| LM Studio | Assistants | API key | From its docs |
| Open WebUI | Assistants | API key | From its docs |
| LibreChat | Assistants | API key | From its docs |
| Perplexity | Assistants | API key | Not ready yet |
| Copilot in JetBrains | Coding tools | API key | From its docs |
| Copilot coding agent | Coding tools | API key | From its docs |
| Junie | Coding tools | API key | From its docs |
| Roo Code | Coding tools | API key | From its docs |
| Kilo Code | Coding tools | API key | From its docs |
| Continue | Coding tools | API key | From its docs |
| OpenCode | Coding tools | API key | Connected |
| Amp | Coding tools | API key | From its docs |
| Warp | Coding tools | API key | From its docs |
| Kiro | Coding tools | API key | From its docs |
| Antigravity | Coding tools | API key | From its docs |
| Qwen Code | Coding tools | API key | Connected |
| Kimi Code | Coding tools | API key | From its docs |
| Amazon Q Developer | Coding tools | API key | From its docs |
| Replit | Coding tools | API key | From its docs |
| Lovable | Coding tools | API key | From its docs |
| goose | Coding tools | Sign-in (OAuth) | From its docs |
| OpenClaw | Coding tools | API key | Not ready yet |
| v0 | Coding tools | API key | Not ready yet |
| JetBrains AI Assistant | Coding tools | API key | Not ready yet |
| Make | Automation | API key | From its docs |
| Claude API | Frameworks | API key | From its docs |
| OpenAI Agents SDK | Frameworks | API key | Tested |
| Vercel AI SDK | Frameworks | API key | Tested |
| CrewAI | Frameworks | API key | Connected |
| Pydantic AI | Frameworks | API key | Tested |
| Mastra | Frameworks | API key | Tested |
| Google ADK | Frameworks | API key | Connected |
| Other clients | Any MCP client | API key | Tested |
Status says how far we checked each one. Tested: we took a screenshot through it. Connected: it added our server and listed its tools. From its docs: we read the setup in the tool's own documentation and have not tried it yet. We never show a setup we did not read there.
The dashboard shows the same setups with your own key already in place, ready to copy.
Any other client
Other clients
For a client that takes a server address and a header. Write your key in place of $LOCALSCREENSHOT_API_KEY.
Address: https://api.localscreenshot.com/mcp Header: Authorization: Bearer $LOCALSCREENSHOT_API_KEY
Tested: we took a screenshot through it.
Then ask
A few more things you can say:
- "Take a full page screenshot of example.com without the cookie banner."
- "Save example.com as a PDF."
- "Show me only the pricing table of example.com/pricing."
- "How many credits do I have left?"
What your AI gets
Three tools:
| Tool | What it does |
|---|---|
take_screenshot |
Takes the screenshot and returns the image, the page title, its final address and status, and the credits left. |
list_countries |
Lists the countries a screenshot can be taken from. |
get_credits |
Says how many credits are left. |
take_screenshot accepts the same options as the API, listed on Take a screenshot, except async and response: the server waits for the picture and decides how to hand it back.
- The image is scaled down to 1,568 pixels wide unless you ask for another
max_width. That keeps it readable for a vision model without spending tokens for nothing. - A PDF is not shown as a picture: your AI gets a short-lived link to download it.
- A very large image (over 4 MB) is not attached either: your AI gets its link.
- An error comes back as a plain sentence with its code, so your AI can tell you what happened. See Errors.
What it costs
The same as the API: 1 credit for a screenshot, 2 from a chosen country. A failed screenshot is not charged. The limits are per account, whichever way you come in.
What it does not do. Your AI cannot use it to sign in to a website, solve a CAPTCHA or get past a paywall. It reads public pages only.