Give your AI agent access to your live Chrome session — works out of the box, connects to tabs you already have open.
pasky/chrome-cdp-skill is ranked 3,563 with 3,279 stars.
Same-category comparison
pasky/chrome-cdp-skill is ranked 3,563 among these 8 peers; google-gemini/computer-use-preview is ranked 3,566.
Same-category comparison| Name | Rank | Metrics |
|---|
| openai/human-eval | 3,545 | - Language:
- Python
- Stars:
- 3,406
- Forks:
- 475
- Description:
- Code for the paper "Evaluating Large Language Models Trained on Code".
|
|---|
| ANative-Lab/EvoAgentX | 3,549 | - Language:
- Python
- Stars:
- 3,384
- Forks:
- 317
- Description:
- 🚀 EvoAgentX: Building a Self-Evolving Ecosystem of AI Agents.
|
|---|
| vladmandic/human | 3,554 | - Language:
- HTML
- Stars:
- 3,351
- Forks:
- 454
- Description:
- Human: AI-powered 3D Face Detection & Rotation Tracking, Face Description & Recognition, Body Pose Tracking, 3D Hand & Finger Tracking, Iris Analysis, Age & Gender & Emotion Prediction, Gaze Tracking, Gesture Recognition.
|
|---|
| vectara/hallucination-leaderboard | 3,556 | - Language:
- Python
- Stars:
- 3,321
- Forks:
- 107
- Description:
- Leaderboard Comparing LLM Performance at Producing Hallucinations when Summarizing Short Documents.
|
|---|
| google-gemini/computer-use-preview | 3,566 | - Language:
- Python
- Stars:
- 3,228
- Forks:
- 407
|
|---|
| blazickjp/arxiv-mcp-server | 3,567 | - Language:
- Python
- Stars:
- 3,221
- Forks:
- 259
- Description:
- A local MCP server for agent literature work. Original-LaTeX section reads, BibTeX from arXiv metadata, and topic watches. Papers stay on disk. Search is optional.
|
|---|
| protectai/llm-guard | 3,570 | - Language:
- Python
- Stars:
- 3,217
- Forks:
- 466
- Description:
- The Security Toolkit for LLM Interactions.
|
|---|
| Chachamaru127/claude-code-harness | 3,573 | - Language:
- Shell
- Stars:
- 3,166
- Forks:
- 302
- Description:
- Claude Code Dedicated Development Harness - Achieving High-Quality Development Through an Autonomous Plan→Work→Review Cycle.
|
|---|
Key parameters and specifications
Key parameters and specifications| Parameter | Value |
|---|
| Language | JavaScript |
|---|
| Category | AI Engineering |
|---|
| Subcategory | Give agent tools, Make agent see hear talk |
|---|
| Keywords | Automation, Data extraction, Tool use |
|---|
| Stars | 3,279 |
|---|
| Forks | 202 |
|---|
| Contributors | 3 |
|---|
| Created date | 2026-03-13 |
|---|
| Updated date | 2026-10-16 |
|---|
| Stars in 1 day | 2 |
|---|
| Stars in 7 days | 7 |
|---|
| Rank | 3,563 |
|---|
| Previous rank | 3,719 |
|---|
| Rank change | 156 |
|---|
| 7-day stars rank | 4,652 |
|---|
| Previous 7-day stars rank | 5,430 |
|---|
| 7-day stars rank change | 778 |
|---|
| Total stars rank | 4,161 |
|---|
| Previous total stars rank | 4,163 |
|---|
| Total stars rank change | 2 |
|---|
| Country | Czech Republic |
|---|
README summary
No browser automation framework, no separate browser instance, no re-login. Works out of the box with any Chrome installation. One toggle to enable, nothing else to install. Why this matters Most browser automation tools launch a fresh, isolated browser. This one connects to the Chrome you're already running, so your agent can: Read pages you're logged into (Gmail, GitHub, internal tools, . ) Interact with tabs you're actively working in See the actual state of a page mid-workflow, not a
clean reload Installation As a pi skill For other agents (Amp, Claude Code, Cursor, etc. ) Clone or copy the directory wherever your agent loads skills or context from. Enable remote debugging in Chrome Navigate to and toggle the switch. The CLI auto-detects Chrome, Chromium, Brave, Edge, and Vivaldi on macOS, Linux, and Windows. If your browser stores in a non-standard location, set the environment variable to the full path. Usage is a unique prefix of the targetId shown
by . chrome-devtools-mcp reconnects on every command, so Chrome's "Allow debugging" modal can re-appear repeatedly and target enumeration times out with many tabs open. On first access to a tab, a lightweight background daemon is spawned that holds the session open. Chrome's "Allow debugging" modal appears once per tab; subsequent commands reuse the daemon silently. Daemons auto-exit after 20 minutes of inactivity. This approach is also why it handles 100+ open tabs reliably, where tools built on
Puppeteer often time out during target enumeration.
Frequently asked questions
How many stars and forks are recorded?
pasky/chrome-cdp-skill has 3,279 stars and 202 forks.
What is the repository rank?
pasky/chrome-cdp-skill is ranked 3,563.
What language and category are recorded?
pasky/chrome-cdp-skill has 3,279 stars, uses JavaScript and belongs to AI Engineering.
Which repository is closest in rank?
pasky/chrome-cdp-skill is compared with google-gemini/computer-use-preview, ranked 3,566 with 3,228 stars.
chrome-cdp
Let your AI agent see and interact with your live Chrome session — the tabs you already have open, your logged-in accounts, your current page state. No browser automation framework, no separate browser instance, no re-login.
Works out of the box with any Chrome installation. One toggle to enable, nothing else to install.
Why this matters
Most browser automation tools launch a fresh, isolated browser. This one connects to the Chrome you're already running, so your agent can:
- Read pages you're logged into (Gmail, GitHub, internal tools, ...)
- Interact with tabs you're actively working in
- See the actual state of a page mid-workflow, not a clean reload
Installation
As a pi skill
pi install git:github.com/pasky/chrome-cdp-skill@v1.0.1
For other agents (Amp, Claude Code, Cursor, etc.)
Clone or copy the skills/chrome-cdp/ directory wherever your agent loads skills or context from. The only runtime dependency is Node.js 22+ — no npm install needed.
Enable remote debugging in Chrome
Navigate to chrome://inspect/#remote-debugging and toggle the switch. That's it.
The CLI auto-detects Chrome, Chromium, Brave, Edge, and Vivaldi on macOS, Linux, and Windows. If your browser stores DevToolsActivePort in a non-standard location, set the CDP_PORT_FILE environment variable to the full path.
Usage
scripts/cdp.mjs list # list open tabs
scripts/cdp.mjs shot # screenshot → runtime dir
scripts/cdp.mjs snap # accessibility tree (compact, semantic)
scripts/cdp.mjs html [".selector"] # full HTML or scoped to CSS selector
scripts/cdp.mjs eval "expression" # evaluate JS in page context
scripts/cdp.mjs nav https://... # navigate and wait for load
scripts/cdp.mjs net # network resource timing
scripts/cdp.mjs click "selector" # click element by CSS selector
scripts/cdp.mjs clickxy # click at CSS pixel coordinates
scripts/cdp.mjs type "text" # type at focused element (works in cross-origin iframes)
scripts/cdp.mjs loadall "selector" # click "load more" until gone
scripts/cdp.mjs evalraw [json] # raw CDP command passthrough
scripts/cdp.mjs open [url] # open new tab (triggers Allow prompt)
scripts/cdp.mjs stop [target] # stop daemon(s)
`` is a unique prefix of the targetId shown by list.
Why not chrome-devtools-mcp?
chrome-devtools-mcp reconnects on every command, so Chrome's "Allow debugging" modal can re-appear repeatedly and target enumeration times out with many tabs open. chrome-cdp holds one persistent daemon per tab — the modal fires once, and it handles 100+ tabs reliably.
How it works
Connects directly to Chrome's remote debugging WebSocket — no Puppeteer, no intermediary. On first access to a tab, a lightweight background daemon is spawned that holds the session open. Chrome's "Allow debugging" modal appears once per tab; subsequent commands reuse the daemon silently. Daemons auto-exit after 20 minutes of inactivity.
This approach is also why it handles 100+ open tabs reliably, where tools built on Puppeteer often time out during target enumeration.