Kaboom-Browser-AI-Devtools-MCP
Health Pass
- License — License: AGPL-3.0
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Community trust — 59 GitHub stars
Code Warn
- fs module — File system access in .github/workflows/dco.yml
Permissions Pass
- Permissions — No dangerous permissions requested
This tool provides browser debugging, inspection, and verification capabilities to AI coding assistants via an MCP server. It bridges local browser development workflows with AI tools by streaming console logs, network failures, and performance data.
Security Assessment
The tool operates inherently with sensitive data, as it intercepts browser network traffic, console outputs, and cookies to relay to your AI assistant. It does not request dangerous broad permissions. A minor warning was flagged for filesystem access, but this is safely isolated to an automated GitHub DCO workflow file rather than the core application. No hardcoded secrets or malicious network requests were detected. Overall risk: Medium. The tool itself is safe, but it handles high-sensitivity local browser data by design, so users should ensure they trust the AI assistant receiving this information.
Quality Assessment
The project is under very active development, with its most recent push occurring today. It is a mature Go application currently on version 0.8.1, featuring cross-platform support (macOS, Linux, Windows) and regular stable releases. The repository is properly licensed under AGPL-3.0. It features solid community engagement with 59 GitHub stars. Additionally, the developers maintain professional standards, utilizing Snyk for security scanning and Codacy for code quality. A clear STABLE/UNSTABLE branch policy ensures developers can reliably pull working code.
Verdict
Use with caution — the codebase itself is well-maintained and secure, but it inherently exposes sensitive browser debugging data to your connected AI tools.
Strum AI DevTools — Browser DevTools for AI coding assistants via MCP
Branch Policy (Read First)
If you want something working, load code and run the server from
STABLE.UNSTABLEmakes zero promises on regressions or issues and is treated as work in progress.
Stable builds are compressed, tagged, and moved toSTABLE.
Kaboom is an AI debugger, inspector, and verification toolkit for local-first browser development workflows. Stream console logs, network failures, exceptions, recordings, and browser evidence into any MCP-compatible coding assistant.
📦 Latest Release
Current version: v0.8.1 — Link health analyzer, browser automation, recording, and performance analysis for AI agents.
macOS / Linux:
curl -sSL https://raw.githubusercontent.com/brennhill/Kaboom-Browser-AI-Devtools-MCP/STABLE/scripts/install.sh | bash
Windows (PowerShell):
irm https://raw.githubusercontent.com/brennhill/Kaboom-Browser-AI-Devtools-MCP/STABLE/scripts/install.ps1 | iex
Quick Start
Fire up Kaboom (binary + extension + auto-config) in one command:
macOS / Linux:
curl -sSL https://raw.githubusercontent.com/brennhill/Kaboom-Browser-AI-Devtools-MCP/STABLE/scripts/install.sh | bash
Windows (PowerShell):
irm https://raw.githubusercontent.com/brennhill/Kaboom-Browser-AI-Devtools-MCP/STABLE/scripts/install.ps1 | iex
This script automatically:
- Downloads the latest stable binary for your platform.
- Installs the browser extension files to
~/.kaboom/extension. - Auto-configures all detected MCP clients (Claude Code, Cursor, Windsurf, Zed, etc.).
Step 1: Finalize Browser Extension
- Open
chrome://extensions - Enable Developer mode (top right)
- Click Load unpacked
- Select the folder:
~/.kaboom/extension(or wherever the script printed)
Step 2: Restart Your AI Tool
Restart Claude Code, Cursor, Windsurf, or Zed. The Kaboom server will now start automatically when needed.
Full setup guide → | Per-tool install guide →
Why Teams Use Kaboom
No debug port required. Other tools need Chrome launched with --remote-debugging-port, which disables security sandboxing and breaks your normal browser workflow. Kaboom uses a standard extension, so your browser stays secure and unmodified.
Single binary, zero runtime. One Go binary that runs anywhere — no runtime dependencies, no Puppeteer, no framework.
Captures what others can't. WebSocket messages, full request/response bodies, user action recording, Web Vitals, automatic regression detection, visual annotations, and Playwright test generation from real browser sessions — features no other MCP browser tool offers.
Works with every MCP tool. Claude Code, Cursor, Windsurf, Zed, Claude Desktop, VS Code + Continue. Switch AI tools without changing your debugging setup.
Enterprise-safe by design. Binds to 127.0.0.1 only. Auth headers are stripped automatically. No accounts, no cloud. Anonymous usage stats only (see Privacy). Audit the source — it's AGPL-3.0.
What It Does
- Console logs —
console.log(),.warn(),.error()with full arguments - Network errors — Failed API calls (4xx, 5xx) with response bodies
- Exceptions — Uncaught errors with full stack traces
- WebSocket events — Connection lifecycle and message payloads
- Network bodies — Request/response payloads for API debugging
- User actions — Click, type, navigate, scroll recording with smart selectors
- Web Vitals — LCP, CLS, INP, FCP with regression detection
- DOM inspection — Query the page with CSS selectors via MCP
- Accessibility audits — WCAG checks with SARIF export
- Security audits — Credentials, PII, headers, cookies, third-party analysis
- Browser automation — Click, type, select, upload, navigate with semantic selectors
- Visual annotations — Draw mode for user feedback with computed style extraction
- Test generation — Playwright tests from context, self-healing selectors, failure classification
- Reproduction scripts — Playwright scripts from recorded user actions
- Noise filtering — Auto-detect and dismiss irrelevant errors
- Developer API —
window.__kaboom.annotate()for custom context
Privacy
All captured data (logs, network, actions) stays 100% local — nothing leaves your machine. No cloud, no accounts.
We collect anonymous usage statistics (feature usage flags, extension version) using a random identifier not linked to your identity. No URLs, browsing data, or personal information is collected.
Performance
See latest benchmarks for current performance data.
Last benchmarked: 2026-02-09 on darwin/arm64 (v0.8.1)
Known Issues
See docs/core/known-issues.md for current known issues.
Development
make test # Go server tests
node --test tests/extension/*.test.js # Extension tests
make dev # Build for current platform
Release process & quality gates → · Changelog →
License
AGPL-3.0 — Free and open source for all use cases.
Artwork, logos, and the Sparky mascot are Copyright (c) Brenn Hill and are not covered by the AGPL. See LICENSE-ARTWORK for details.
Fueling rapid development with AI
If you find Kaboom useful, please consider giving it a star.
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found