Webpage ScreenshotSAFE
An MCP that gives your agent the ability to snap a screenshot of webpages. Useful when you want your agent to check its progress during development.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
An MCP (Model Context Protocol) server that captures screenshots of web pages using Puppeteer. This server allows AI agents to visually verify web applications and see their progress when generating web apps.
Features
- Full page screenshots: Capture entire web pages or just the viewport
- Element screenshots: Target specific elements using CSS selectors
- Multiple formats: Support for PNG, JPEG, and WebP formats
- Customizable options: Set viewport size, image quality, wait conditions, and delays
- Base64 encoding: Returns screenshots as base64 encoded images for easy integration
- Authentication support: Manual login and cookie persistence
- Default browser integration: Use your system's default browser for a more natural experience
- Session persistence: Keep browser sessions open for multi-step workflows
Installation
Quick Start (Claude Desktop Extension)
Drag and drop the generated screenshot-webpage-mcp.dxt file into Claude Desktop for automatic installation!
Manual Installation
To install and build the MCP from source:
# Clone the repository (if you haven't already) git clone https://github.com/ananddtyagi/webpage-screenshot-mcp.git cd webpage-screenshot-mcp # Install dependencies npm install # Build the project npm run build
The MCP server is built using TypeScript and compiled to JavaScript. The dist folder contains the compiled JavaScript files.
Adding to Claude or Cursor
To add this MCP to Claude Desktop or Cursor:
- Claude Desktop:
- Go to Settings > Developer
- Click "Edit Config"
- Add the following:
"webpage-screenshot": {
"command": "node",
"args": [
"~/path/to/webpage-screenshot-mcp/dist/index.js"
]
}- Save and reload Claude
- Cursor:
- Open
3a6207b3e2f2OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add webpage-screenshot-mcp -- npx -y [email protected]
{
"mcpServers": {
"webpage-screenshot-mcp": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (5)
4 read · 0 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
clear-auth-cookies | destructive | Clears saved authentication cookies for a specific domain or all domains |
login-and-wait | read | Opens a webpage in a visible browser window for manual login, waits for user to complete login, then saves cookies |
screenshot-element | read | Captures a screenshot of a specific element on a webpage using a CSS selector |
screenshot-page | read | Captures a screenshot of a given URL and returns it as base64 encoded image. Can use saved cookies from login-and-wait. |
signal-login-complete | read | Signals that manual login is complete and the login-and-wait tool should continue |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- none-observed
- Shell
- declared (4 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (3)
clear-auth-cookies
@modelcontextprotocol/sdk, zod, @types/node, typescript
screenshot-webpage-mcp.dxt
Gates applied: no_behavioural_pass.
3a6207b3e2f2full audit observations/trust-audit/mcp-server/ananddtyagi__webpage-screenshot.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 3a6207b3e2f2 | SAFE | B | 89 | first audit |
Questions
What is the Webpage Screenshot MCP server?
An MCP that gives your agent the ability to snap a screenshot of webpages. Useful when you want your agent to check its progress during development.
What tools does Webpage Screenshot expose?
5 in total: 4 read-only, 0 that write, and 1 that can delete or overwrite (clear-auth-cookies). Every one is listed on this page with its risk.
Is Webpage Screenshot safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Webpage Screenshot need?
No credential environment variables were found in its source, so it appears to need none.
How does Webpage Screenshot run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as webpage-screenshot-mcp at 1.0.0.
How current is this page?
The grade is for one exact copy of the source (3a6207b3e2f2), read on 2026-10-08. The repository is watched and re-audited when it changes.