Which integrations are available for this server?

Enables capturing screenshots and motion from Google Maps, displaying rendered map tiles, overlays, and any blocking elements like consent modals. Enables capturing visual evidence of Instagram pages, including rendered state, consent walls, login overlays, and motion analysis. Allows capturing visuals from TikTok pages, including rendered video players, consent dialogs, and motion analysis. Provides ability to capture screenshots and motion evidence from YouTube pages, showing actual rendered video player, overlays, and consent walls.

How do I use RenderProof?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@RenderProof Capture a full-page screenshot of https://example.com" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

RenderProof

by Djsand

Overview Schema Related Servers Score Discussions

TypeScript

Local

RenderProof

Rendered web evidence for coding agents.

RenderProof is a local-first MCP server that lets Codex, Claude Code, Cursor, and other agents capture what Chromium actually rendered: screenshots, motion, CSS animation metadata, pixel diffs, gates, loading states, and visual evidence that plain fetch tools miss.

It is not a scraper. It is not a bypass tool. It is the receipt for what the browser saw.

Why This Exists

Coding agents are very good at reading source. The web is increasingly bad at being understood from source alone.

Modern pages hide their real state behind client rendering, consent walls, cookie modals, skeleton loaders, canvas, video surfaces, iframes, maps, bot checks, and motion. A text fetch can say "Google Maps"; a browser screenshot can show "Danish consent wall blocking the map."

RenderProof gives agents a grounded visual evidence layer before they make claims or copy a design.

Related MCP server: Snowfort Circuit MCP

What It Does

Tool	Use it when	Output
`route_web_task`	The agent needs to choose text, screenshot, motion, search, or another platform tool	Route recommendation with evidence
`read_url`	Text is enough	Normalized direct fetch text, with optional remote reader
`capture_page`	Layout, gates, rendered state, canvas, images, maps, PDFs-as-browser-state, or visual QA matter	PNG screenshot + metadata
`capture_motion`	A human needs to see animation, loading, scroll, transitions, or media motion	WebM video + keyframe PNGs
`analyze_motion`	A coding agent needs to recreate or critique motion	Contact sheet, diff image, CSS animation metadata, pixel-diff JSON, design notes
`doctor`	You want to verify local runtime/browser setup	Runtime and Playwright checks

The Core Idea

flowchart LR
  A["Agent task"] --> B["route_web_task"]
  B --> C["read_url"]
  B --> D["capture_page"]
  B --> E["capture_motion"]
  B --> F["analyze_motion"]
  C --> G["Grounded answer"]
  D --> G
  E --> G
  F --> G

RenderProof makes the expensive path explicit: read text first when text is enough, escalate to rendered evidence when pixels matter.

Example Prompts

Use RenderProof to capture a full-page screenshot of https://example.com with autoScrollBeforeCapture true.

Use RenderProof to analyze the animation style on this page and return the contact sheet, pixel-diff summary, and CSS animation metadata.

Use RenderProof to check what the browser actually sees on this Instagram/TikTok/YouTube page.

Use RenderProof to capture visual evidence before claiming whether this page is blocked by a consent wall, login wall, CAPTCHA, or skeleton loader.

Motion Analysis

Video is useful for humans. Agents need structured evidence.

analyze_motion samples frames over time and returns:

a contact sheet PNG
a red-highlight diff PNG
frame timestamps
changed-pixel ratios
changed-region bounds and centroid movement
dominant movement direction
document.getAnimations() metadata
CSS keyframes, duration, easing, direction, iterations, target selector, animated properties
design notes written for a coding agent

Example design notes:

[
  "Detected 2 CSS/Web Animation target(s), affecting transform.",
  "Primary timing appears to use duration 1200ms, easing ease-in-out, direction alternate.",
  "Pixel sampling saw up to 5.81% of sampled pixels change between adjacent frames.",
  "The changed region centroid trends right across sampled frames."
]

That is the difference between "there is an animation" and "recreate this with transform, 1200ms, ease-in-out, alternate, moving right while scaling."

Install

One-liner:

curl -fsSL https://raw.githubusercontent.com/Djsand/renderproof/main/install.sh | bash -s -- codex

Windows PowerShell:

powershell -NoProfile -ExecutionPolicy Bypass -Command "& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Djsand/renderproof/main/install.ps1))) -Target codex"

Swap codex for claude, cursor, cline, windsurf, gemini, or generic.

The codex target writes directly to ~/.codex/config.toml, so it works even when the Codex CLI is not available in your terminal PATH.

If Windows says Missing required command: node, install Node.js 20+ and reopen PowerShell:

winget install OpenJS.NodeJS.LTS

Manual install:

git clone https://github.com/Djsand/renderproof.git
cd renderproof
npm install
npx playwright install chromium

Then install it into your coding agent:

node dist/index.js install

One-liners:

curl -fsSL https://raw.githubusercontent.com/Djsand/renderproof/main/install.sh | bash -s -- codex
curl -fsSL https://raw.githubusercontent.com/Djsand/renderproof/main/install.sh | bash -s -- claude
curl -fsSL https://raw.githubusercontent.com/Djsand/renderproof/main/install.sh | bash -s -- cursor
curl -fsSL https://raw.githubusercontent.com/Djsand/renderproof/main/install.sh | bash -s -- cline
curl -fsSL https://raw.githubusercontent.com/Djsand/renderproof/main/install.sh | bash -s -- windsurf
curl -fsSL https://raw.githubusercontent.com/Djsand/renderproof/main/install.sh | bash -s -- gemini

Windows PowerShell one-liners:

powershell -NoProfile -ExecutionPolicy Bypass -Command "& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Djsand/renderproof/main/install.ps1))) -Target codex"
powershell -NoProfile -ExecutionPolicy Bypass -Command "& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Djsand/renderproof/main/install.ps1))) -Target claude"
powershell -NoProfile -ExecutionPolicy Bypass -Command "& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Djsand/renderproof/main/install.ps1))) -Target cursor"
powershell -NoProfile -ExecutionPolicy Bypass -Command "& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Djsand/renderproof/main/install.ps1))) -Target cline"
powershell -NoProfile -ExecutionPolicy Bypass -Command "& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Djsand/renderproof/main/install.ps1))) -Target windsurf"
powershell -NoProfile -ExecutionPolicy Bypass -Command "& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Djsand/renderproof/main/install.ps1))) -Target gemini"

Local helper equivalents:

node dist/index.js install codex --write-user
node dist/index.js install claude --apply
node dist/index.js install cursor --write-project
node dist/index.js install cline --write-user
node dist/index.js install windsurf --write-user
node dist/index.js install gemini --apply --scope user

See docs/installation.md for agent-specific setup.

Use With Codex

node dist/index.js install codex --write-user

Manual equivalent:

codex mcp add renderproof -- node /absolute/path/to/renderproof/dist/index.js mcp

Direct config equivalent:

[mcp_servers.renderproof]
command = "node"
args = ["/absolute/path/to/renderproof/dist/index.js", "mcp"]

Use With Claude Code

node dist/index.js install claude --apply

Manual equivalent:

claude mcp add renderproof -- node /absolute/path/to/renderproof/dist/index.js mcp

Use As CLI

npm run dev -- route "summarize this page" --url https://example.com
npm run dev -- read https://example.com
npm run dev -- capture https://example.com --full-page
npm run dev -- capture https://example.com --full-page --auto-scroll
npm run dev -- motion https://example.com --duration 5000 --keyframes true
npm run dev -- analyze-motion https://example.com --duration 3000 --samples 5
npm run dev -- install generic --json
npm run doctor

Safety Model

RenderProof is designed to show gates, not bypass them.

By default, it:

allows public http and https URLs
blocks localhost, private IP ranges, and hosts resolving to private IPs
does not use remote readers unless explicitly enabled
does not solve CAPTCHA
does not hide automation
does not click through consent, login, or payment walls
returns structured failures instead of pretending a page was accessible

Environment controls:

RENDERPROOF_ALLOWED_HOSTS=example.com,*.example.org
RENDERPROOF_ALLOW_PRIVATE_NETWORK=1
RENDERPROOF_ENABLE_REMOTE_READERS=1
RENDERPROOF_OUTPUT_DIR=.renderproof/evidence
RENDERPROOF_MAX_CHARS=12000
JINA_API_KEY=...

Legacy GROUNDED_WEB_* variables are still supported for compatibility.

What RenderProof Is Good At

Seeing rendered browser state when direct fetch returns useless HTML
Capturing cookie modals, login walls, consent walls, reCAPTCHA widgets, and skeleton loaders
Capturing maps, video players, image pages, canvas/WebGL surfaces, and JS-computed UI
Producing full-page screenshots with optional pre-scroll for lazy loading
Recording short WebM motion evidence for humans
Turning animation into agent-readable CSS and pixel-diff evidence
Creating auditable local evidence files for coding workflows

What It Will Not Do

bypass paywalls, login walls, CAPTCHA, bot checks, or private networks by default
scrape social platforms as a data product
replace Playwright MCP for interactive browser control
guarantee semantic object tracking from pixel diffs alone

Playwright is the engine. RenderProof is the evidence layer.

Development

npm install
npm run typecheck
npm run build
npm run doctor

Run a browser launch check:

npm run dev -- doctor --check-browser-launch

Roadmap

PDF rendering and text extraction
page-state detection for consent/login/CAPTCHA/skeleton/download/blank-page states
style-reference capture for "build me a site like this" workflows
safe user-provided auth/session support
evidence bundle manifests
optional interaction sequences with before/after evidence

License

MIT

This server cannot be installed

license - permissive license

quality - not tested

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

–Release cycle

–Releases (12mo)

Commit activity

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

Lightport: Open-Sourcing Glama's AI Gateway
By punkpeye on April 27, 2026.
open source
OpenAI
Tool Definition Quality Score (TDQS)
By punkpeye on April 3, 2026.
mcp
The Hackers Who Tracked My Sleep Cycle
By punkpeye on March 26, 2026.
security

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Djsand/renderproof'

If you have feedback or need assistance with the MCP directory API, please join our Discord server