What is chatgpt-mcp-browser-debugger?

A debugging guide for browser automation that controls ChatGPT through Chrome’s debugging interface and Playwright scripts. It separates live browser inspection from repeatable repository tests.

Use it to inspect the live ChatGPT browser, verify browser interactions, run diagnostic scripts, and investigate automation failures.

It provides a defined way to investigate stuck menus, failed chat creation, and incorrect browser selections before changing automation code.

How trankhanh040147/chatgpt-mcp configures Cursor

Reuse

Borrowing it

Nothing to install: this file belongs to trankhanh040147/chatgpt-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/trankhanh040147/chatgpt-mcp/main/.cursor/skills/chatgpt-mcp-browser-debugger/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/trankhanh040147/chatgpt-mcp

Made for: Cursor.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/trankhanh040147/chatgpt-mcp/chatgpt-mcp-browser-debugger/github.svg)](https://agentmods.dev/skills/trankhanh040147/chatgpt-mcp/chatgpt-mcp-browser-debugger)

Your own site

<a href="https://agentmods.dev/skills/trankhanh040147/chatgpt-mcp/chatgpt-mcp-browser-debugger"><img src="https://agentmods.dev/badge/skills/trankhanh040147/chatgpt-mcp/chatgpt-mcp-browser-debugger/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/trankhanh040147/chatgpt-mcp/chatgpt-mcp-browser-debugger"><img src="https://agentmods.dev/badge/skills/trankhanh040147/chatgpt-mcp/chatgpt-mcp-browser-debugger.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 79 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,591 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00079	$0.01591
Opus 5	$0.00039	$0.00796
Sonnet 5	$0.00016	$0.00318
Haiku 4.5	$0.00008	$0.00159

Measured 10d ago against content hash 93949c25f46e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

chatgpt-mcp-browser-debugger scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s http://127.0.0.1:9222/json/version | head -1

.cursor/skills/chatgpt-mcp-browser-debugger/SKILL.md · 153 lines

How it starts

The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.

chatgpt-mcp-browser-debugger

Development/debug plane for chatgpt-mcp worker browser automation. Production remains browser-worker → Playwright connectOverCDP → Chrome :9222. MCP does not replace production.

Prerequisites

Env: /Users/vulcanlabs/src/gh/obsidian-vault/vault-mac-1/configs/chatgpt-mcp-env.md
CDP Chrome logged in (profile $HOME/chrome-chatgpt-debug, not Default Chrome)
Stack up: npm run build && node dist/gptmcp.js restart (or ./scripts/start-broker-stack.sh)
Chrome DevTools MCP in .cursor/mcp.json → chrome-devtools with --browser-url=http://127.0.0.1:9222
Reload MCP in Cursor after config change

Verify:

curl -s http://127.0.0.1:9222/json/version | head -1
curl -s http://127.0.0.1:8787/health

Two layers (do not collapse)

Layer	Tool	Role
Exploration	Chrome DevTools MCP (`take_snapshot`, `take_screenshot`, `click`, `evaluate_script`)	See real DOM/a11y, disambiguate nodes, prove interaction
Regression	Repo scripts (`diagnose-plus-menu`, `e2e-create-chat`)	Deterministic exit code, JSON artifacts, CI-adjacent repeatability

MCP finds the truth. Scripts encode the truth.

Autonomous debug loop

START
  gptmcp status / doctor
       │
  verify :9222 and :18788 (broker)
       │
  Chrome DevTools MCP ──► select ChatGPT worker tab (Chat surface)
       │
  snapshot BEFORE action
       │
  ONE action (e.g. click composer-plus-btn)
       │
  snapshot + screenshot AFTER
       │
  Expected UI state?
    YES ──► encode locator in create-chat.ts
    NO  ──► inspect a11y tree, bbox, console; do NOT patch selector from screenshot alone
       │
  npm run build
       │
  node dist/gptmcp.js restart
       │
  npm run e2e:create-chat  (or dashboard New chat)
       │
  LIVE RETEST ──► loop until pass

Chrome DevTools MCP workflow (composer + → Cursor)

Do not modify code until steps 1–9 succeed via MCP.

Attach to existing Chrome: --browser-url=http://127.0.0.1:9222
List pages / select active ChatGPT tab (chatgpt.com, Chat not Work if possible)
take_snapshot — record UIDs
Find composer-plus-btn (or equivalent in a11y tree)
Click +
take_snapshot + take_screenshot — menu open
Find every visible node containing text Cursor
For each: note UID, parent chain, position — separate sidebar Plugins vs composer + menu
Click the composer-menu Cursor (anchor discovery to composer-plus-btn / composer bbox, not global getByText)
take_snapshot — verify Cursor chip on composer (not sidebar link)
Only then edit src/browser/create-chat.ts to match proven interaction
Run npm run diagnose:plus-menu and npm run e2e:create-chat to regression-lock

Read the full file on GitHub · 153 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

10d ago First seen · 153 lines · 79 tokens per session scan A 93949c25f46e

Subscribe to this mod's changes

chatgpt-mcp-browser-debugger is a skill published in the GitHub repository trankhanh040147/chatgpt-mcp (0 stars, last pushed 5d ago), licensed MIT. It adds 79 tokens to every session and 1,591 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Browser automation Debugging

chrome-devtools-cli

Use this skill to write shell scripts or run shell commands to automate tasks in the browser or otherwise use Chrome DevTools via CLI.

ChromeDevTools/chrome-devtools-mcp · 31 tokens

a11y-debugging

Uses Chrome DevTools MCP for accessibility (a11y) debugging and auditing based on web.dev guidelines. Use when testing semantic HTML, ARIA labels, focus states, keyboard navigation, tap targets, and color contrast.

ChromeDevTools/chrome-devtools-mcp · 50 tokens

chrome-devtools

Uses Chrome DevTools via MCP for efficient debugging, troubleshooting and browser automation. Use when debugging web pages, automating browser interactions, analyzing performance, or inspecting network requests. This skill does not apply to --slim mode (MCP configuration).

ChromeDevTools/chrome-devtools-mcp · 55 tokens

chrome-performance-optimizer

Autonomous multi-agent performance optimization loop for Chromium and V8. Supports profile-seeded mode (analyzing Speedometer 3 / JetStream profiles via pprof or Sagacity MCP) and pattern-driven discovery mode (fan-out exploration of Blink and V8 macro-patterns grounded in historical wins and past rejected CLs).…

chromium/chromium · 119 tokens

nullaway

Guide for resolving NullAway static analysis errors. Best practices for: Passing ObservableSupplier/Supplier Dereferencing potentially @Nullable values Adding @NullMarked to Java code.

chromium/chromium · 43 tokens

bisect

Bisect regressions or broken builds in the Chromium repository using deterministic binary search, checking out midpoint commits, syncing dependencies with gclient sync, compiling the target with autoninja, and querying the user for test results until the culprit commit is isolated.

chromium/chromium · 52 tokens

chatgpt-mcp: Skill for Cursor