brennhill/kaboom-browser-ai-devtools-mcp/kaboom

Claude Code and agent skill by brennhill/kaboom-browser-ai-devtools-mcp

Trust tier 2
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Use when user asks to check browser state, debug page errors, inspect network traffic, take screenshots, automate clicks or form fills, run accessibility or security audits, measure performance, generate tests, or record browser sessions. Trigger phrases: "check my browser", "debug this page", "take a screenshot", "browser errors", "network requests", "automate the browser", "record a test", "accessibility audit", "performance check", "what's on the page".

Install

npx vskill@latest install brennhill/kaboom-browser-ai-devtools-mcp/kaboom

SKILL.md

Kaboom

Critical: Daemon Setup

Before any tool call, ensure the daemon is running:

bash
bash scripts/ensure-daemon.sh

If this fails, run the connection doctor:

bash
bash scripts/connection-doctor.sh

How to Call Tools

All 5 tools are called via the helper script:

bash
bash scripts/kaboom-call.sh <tool> '<json_arguments>'

Examples:

bash
# Read browser errors
bash scripts/kaboom-call.sh observe '{"what":"errors","limit":20}'
# Take a screenshot
bash scripts/kaboom-call.sh observe '{"what":"screenshot","format":"png"}'
# Click an element
bash scripts/kaboom-call.sh interact '{"what":"click","selector":"button.submit"}'
# Run accessibility audit
bash scripts/kaboom-call.sh analyze '{"what":"accessibility"}'
# Generate a Playwright test
bash scripts/kaboom-call.sh generate '{"what":"test","test_name":"login_flow"}'
# Check health
bash scripts/kaboom-call.sh configure '{"what":"health"}'

Tools Quick Reference

observe — Read captured browser telemetry (passive)

Top modes: errors, logs, network_waterfall, network_bodies, actions, screenshot, page, tabs, storage, recordings, websocket_events, error_bundles, timeline, transients, page_inventory

Common params: what (required), limit, since_cursor, summary, format, url, scope

Pagination: pass after_cursor/before_cursor/since_cursor from response metadata. Set restart_on_eviction=true if cursor expired.

Full reference: references/observe-modes.md

analyze — Active analysis queries (sent to extension)

Top modes: dom, accessibility, security_audit, performance, forms, link_health, page_summary, computed_styles, visual_diff, api_validation, third_party_audit, audit

Common params: what (required), selector, timeout_ms, background, summary

Async mode: set background:true to get correlation_id, then poll with observe(what='command_result', correlation_id=...).

Full reference: references/analyze-modes.md

generate — Create artifacts from captured data

Top modes: test, reproduction, pr_summary, har, csp, sarif, visual_test, test_from_context, test_heal, test_classify

Common params: what (required), test_name, save_to, last_n

Full reference: references/generate-modes.md

configure — Session management and diagnostics

Top modes: health, doctor, store, clear, describe_capabilities, event_recording_start, event_recording_stop, playback, noise_rule, streaming, telemetry, diff_sessions

Common params: what (required), store_action, key, value

Full reference: references/configure-modes.md

interact — Browser automation

Top modes: click, type, navigate, screenshot, execute_js, explore_page, list_interactive, fill_form, scroll_to, hover, wait_for, batch

Common params: what (required), selector, text, url, script

Targeting priority: selector > element_id > index > x/y coordinates

Semantic selectors: text=Submit, role=button, placeholder=Email

Composable enrichment: add include_screenshot:true, action_diff:true, or wait_for_stable:true to any action.

Full reference: references/interact-actions.md

Workflows

Choose the workflow that matches your task:

TaskWorkflow
Debug errors, triage failures, create regression testsreferences/workflow-debug-and-triage.md
Site audit, UX review, accessibility, security, API validationreferences/workflow-audit-and-security.md
Browser automation, demos, test generation, release gatesreferences/workflow-automate-and-test.md
Performance measurement and comparisonreferences/workflow-performance.md

Troubleshooting

Daemon not running

Run: bash scripts/ensure-daemon.sh If fails: check which kaboom — binary must be on PATH. Install: npm install -g kaboom-agentic-browser

Extension not connected

Run: bash scripts/connection-doctor.sh Check Chrome extension is installed and enabled on the target page. The extension POSTs telemetry to localhost:7890.

No data returned

Ensure a tab is tracked:

bash
bash scripts/kaboom-call.sh observe '{"what":"tabs"}'

If no tracked tab, open a page with the extension active.

Tool call returns error

Use describe_capabilities for parameter help:

bash
bash scripts/kaboom-call.sh configure '{"what":"describe_capabilities","tool":"observe","mode":"errors"}'
Connection lost mid-session

Re-run ensure-daemon and retry:

bash
bash scripts/ensure-daemon.sh
bash scripts/kaboom-call.sh configure '{"what":"health"}'

Security

  • Security scanPassed, 93 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
65
Repository forks
4
7-day trend score
21

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/brennhill/kaboom-browser-ai-devtools-mcp/kaboom/badge)](https://verified-skill.com/skills/brennhill/kaboom-browser-ai-devtools-mcp/kaboom)

Something wrong with this skill or its page? Tell us.