Automated summary
Ghost Chrome is a high-stealth, headful browser automation service running in Docker, designed to bypass anti-bot protections like Cloudflare by simulating human interaction with a real Chrome instance.
Strengths
- Effectively bypasses advanced anti-bot protections
- Simulates realistic human interactions (Bezier curves, variable delays)
- Uses a real Chrome instance, unlike headless approaches
- Easy to use via simple terminal commands
Limitations
- Requires Docker and prior system setup
- Can be slower than standard headless browsers
- Service updates may require manual intervention
Use Ghost Chrome when you need to automate tasks on websites that block headless browsers or employ strong anti-bot detection (Cloudflare, Google, etc.).
Do not use it for simple scraping or standard automation tasks where a regular headless browser (Playwright, Puppeteer) is sufficient.
Security analysis
SafeThe skill exposes a controlled set of bash commands (all prefixed 'ghost-browser') that run a Docker-based browser automation tool. No destructive commands (e.g., rm -rf, curl | sh), exfiltration, or disabling of safety features are present. While the tool could be used for unauthorized scraping or bypassing protections, the skill itself does not instruct or directly enable those actions beyond declaring the tool's capability.
No concerns found
How this analysis is madeExamples
Log into example.com/login using Ghost Chrome. Start the browser, open the page, find the email input and submit button using interactive snapshot, fill the email field with 'user@test.com', click submit, and wait for the dashboard to load.I need to scrape content from a website that blocks headless browsers with Cloudflare. Use Ghost Chrome to start a stealth browser session, navigate to the target URL, take a snapshot of the page text, and save a screenshot to verify.Do a Google search for 'best AI tools 2025' using Ghost Chrome to avoid bot detection. Open google.com, type the query into the search box, press Enter, wait for results, and then take a screenshot of the search results page.name: Ghost Chrome description: A high-stealth, headful browser automation service running in Docker. Use this SPECIFICALLY when standard headless browsers (agent-browser/Playwright) are blocked by anti-bot protections (Cloudflare, Google, etc.). It runs a real Chrome instance and simulates human input. read_when:
- Bypassing strong bot detection
- Automating sites that block headless browsers
- Performing Google searches or interactions requiring high trust metadata: {"clawdbot":{"emoji":"👻","requires":{"bins":["docker","curl","jq"]}}} allowed-tools: Bash(ghost-browser:*)
Stealth Browser Automation with Ghost Chrome
Core Workflow
- Always Start First: Run
ghost-browser startat the beginning of your task. It is fast and safe to run even if already running. - Navigate & Act: Use standard commands.
- Finish: You can explicitly run
ghost-browser stopor let it auto-shutdown after 5 minutes of inactivity.
Commands
Lifecycle
ghost-browser start # Boot the Docker container (Safe to run repeatedly)
ghost-browser stop # Stop and remove the container
ghost-browser restart # Restart the service (get a fresh browser session)
ghost-browser update # Update from git and rebuild the environment
ghost-browser status # Check service health
Updating to Latest
If you need to update to the latest version of the toolkit:
ghost-browser update # Automatically pull code, install dependencies, and rebuild the browser container
Navigation & Inspection
ghost-browser open <url> # Navigate to URL (waits for network idle) (alias: navigate)
ghost-browser snapshot # Get page text content
ghost-browser snapshot -i # Get interactive elements with refs (@e1, @e2)
ghost-browser screenshot # Save screenshot to screenshot.png
ghost-browser highlight <sel> # Draw outline around element (visual check)
ghost-browser tabs # List open tabs/windows
ghost-browser switch <index> # Switch focus to a specific tab
ghost-browser frame <sel> # Focus an iframe ('main' to return to top)
ghost-browser cookies # View or set (ghost-browser cookies set name val)
Interaction (Stealth)
All interactions use human simulation algorithms (Bezier curves, variable delays). You can use CSS selectors OR references from snapshot -i (e.g., @e1).
ghost-browser click <selector> # Click an element
ghost-browser dblclick <selector> # Double-click an element
ghost-browser drag <src> <dest> # Drag one element to another
ghost-browser hover <selector> # Move mouse to element
ghost-browser type <selector> <text> # Focus and type text with jitter
ghost-browser fill <selector> <text> # Clear field and type text
ghost-browser scroll <y_offset> # Humanized scroll (positive = down)
ghost-browser press <key> # Press key (Enter, Escape, Tab, etc.)
ghost-browser upload <sel> <path> # Upload a file
ghost-browser wait <selector> # Wait for element to appear
Example: Form Submission with Refs
# 1. Open and find elements
ghost-browser start
ghost-browser open "https://example.com/login"
ghost-browser snapshot -i
# Output: [{"id": "@e1", "tagName": "input", "text": "Email"}, {"id": "@e2", "tagName": "button", "text": "Submit"}]
# 2. Interact using refs
ghost-browser fill @e1 "user@test.com"
ghost-browser click @e2
ghost-browser wait ".dashboard"
Troubleshooting
- "Element not found": The selector might be wrong, or the page hasn't loaded. Use
ghost-browser screenshotorghost-browser viewto check. - Service failed to start: Check
docker logs ghost-browser.
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/stealth-browser-automation-with-ghost-chrome-mm8o527u/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/stealth-browser-automation-with-ghost-chrome-mm8o527u' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/stealth-browser-automation-with-ghost-chrome-mm8o527u/download' --output '.skill-downloads/stealth-browser-automation-with-ghost-chrome-mm8o527u/SKILL.md.part' && mv '.skill-downloads/stealth-browser-automation-with-ghost-chrome-mm8o527u/SKILL.md.part' '.skill-downloads/stealth-browser-automation-with-ghost-chrome-mm8o527u/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
Tools declared by the author (check availability in your agent): Bash(ghost-browser:*)
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
This file does not provide a usable name/description header for this native procedure. It remains available to download and inspect; use the manual method below.
Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/stealth-browser-automation-with-ghost-chrome-mm8o527u/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
ClickHouse Performance Comparison
Testing
Evaluate ClickHouse performance test results, check PR performance changes, compare with history, and assess whether a result is actionable.
Analyze CI Performance Reports
Testing
Analyze CI performance comparison reports for a ClickHouse PR. Cross-references master history to distinguish real regressions from flaky tests.
Santa Method
Testing
Multi-agent adversarial verification with two independent reviewers. Output is checked twice and iteratively fixed until both pass before shipping.