Browser AI Debate MCP
Integrates with Gemini Web to allow AI debate by sending questions, receiving answers, and facilitating multi-round discussions through browser automation.
Integrates with ChatGPT Web to allow AI debate by sending questions, receiving answers, and facilitating multi-round discussions through browser automation.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Browser AI Debate MCPDebate the pros and cons of using TypeScript for a large project"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Browser AI Debate MCP
A Model Context Protocol (MCP) server that runs a structured multi-round debate between ChatGPT Web and Gemini Web by operating both in Chrome via CDP (Chrome DevTools Protocol). It uses browser automation rather than provider APIs.
What It Does
Opens ChatGPT Web and Gemini Web in the same Chrome session
Sends the same question to both models independently
Makes each model critique the other response
Requests revised answers
Produces a final conclusion and an independent audit
Returns structured results with confidence, rationale, risks, and provider status
Related MCP server: guru-pk-mcp
Important
Requires logged-in ChatGPT and Gemini browser sessions
Uses browser UI automation, so provider UI changes can break selectors
Does not require OpenAI or Gemini API keys
Stores the dedicated profile only in
.browser-ai-profile/, which is gitignoredDoes not terminate a user-owned Chrome process
Prerequisites
Node.js 20 or newer
Google Chrome desktop
Setup
git clone https://github.com/kaenozu/browser-ai-debate-mcp.git
cd browser-ai-debate-mcp
npm install
npm run buildChrome Preparation
Auto-launch
When BROWSER_AI_CDP_URL is not set, the server attempts to launch a dedicated Chrome profile automatically.
npm startConnect to an existing logged-in Chrome
Log into ChatGPT and Gemini in Chrome.
Enable remote debugging at
chrome://inspect/#remote-debugging.Configure the CDP endpoint before starting the server.
Windows Command Prompt:
set BROWSER_AI_CDP_URL=http://127.0.0.1:9222
npm startPowerShell:
$env:BROWSER_AI_CDP_URL = "http://127.0.0.1:9222"
npm startPre-launch a dedicated profile
npm run loginLog into both services, close Chrome, then run:
npm run chrome:debugStart the MCP server in another terminal:
npm startMCP Client Configuration
The public server identifier remains browser-ai-debate, and the public tool name remains browser_ai_debate for compatibility.
Auto-launch configuration:
{
"mcpServers": {
"browser-ai-debate": {
"command": "node",
"args": ["dist/index.js"]
}
}
}Existing Chrome via CDP:
{
"mcpServers": {
"browser-ai-debate": {
"command": "node",
"args": ["dist/index.js"],
"env": {
"BROWSER_AI_CDP_URL": "http://127.0.0.1:9222"
}
}
}
}browser_ai_debate Tool
Example input:
{
"question": "Why is not swallowing exceptions important in TypeScript?",
"context": "We have 70 source files and 23 try-catch blocks that swallow errors",
"decisionCriteria": ["reliability", "debuggability", "user experience", "maintainability"],
"maxRounds": 3,
"timeoutMs": 900000
}Field | Required | Default | Description |
| Yes | — | Debate question, up to 100,000 characters |
| No | — | Additional context |
| No | — | Evaluation criteria |
| No | 3 | Maximum 1–5 rounds |
| No | 900000 | Total timeout, minimum 30,000 ms |
| No | derived | Per-provider timeout |
| No | true | Include the transcript |
| No |
| Provider that produces the final decision |
Timeout and Partial Results
timeoutMscovers queue wait time and all debate phasesProvider failures are returned explicitly
If only one provider completes, the result is marked
partialPartial output is never presented as a complete result
Security Notes
Browser profiles and credentials are not committed to Git
Conversation content is not sent to an additional external service by this MCP server
User-owned Chrome is not terminated by the server
Dedicated Chrome processes started by the server are cleaned up on shutdown
Troubleshooting
CDP connection fails
Confirm Chrome remote debugging is enabled.
Confirm
BROWSER_AI_CDP_URLpoints to the active endpoint.Check that port 9222 is available and reachable.
Run
npm run chrome:debugto start the dedicated profile explicitly.Open
http://127.0.0.1:9222/json/versionlocally to confirm the endpoint responds.
AUTH_REQUIRED
The stored browser session has expired or is not logged into one of the providers.
Run
npm run login.Log into ChatGPT and Gemini.
Close the profile window and retry.
Provider UI or DOM selectors changed
Authentication succeeds, but the provider page no longer matches the expected UI.
Retry with a fresh dedicated profile.
Confirm the provider page works manually.
Report the provider, page state, and observed error without including credentials or private conversation content.
Update only the provider selector module (
src/providers/<provider>/selectors.ts) and its sanitized fixture test (tests/provider-selector-smoke.test.ts). Keep fixtures free of real conversation content.Run
npm testandnpm run checkbefore opening a PR.
The normal CI suite runs the sanitized DOM fixture and fallback-order tests. The opt-in .github/workflows/live-smoke.yml is manual-only, checks out master, requires the protected provider-live-smoke environment, and reads only BROWSER_AI_CDP_URL from secrets. It must never be enabled for pull requests or execute untrusted PR code. Live smoke requires logged-in ChatGPT and Gemini tabs in the operator's CDP session; those checks are intentionally an external gate.
Auto-launch cannot find Chrome
Confirm Chrome is installed in a standard location, or use an explicit CDP connection with BROWSER_AI_CDP_URL.
Windows Quick Start
start.batManual start:
npm run build
npm startInteractive MCP inspection:
npm run inspectorOpenCode MCP Registration
Existing logged-in Chrome:
{
"mcpServers": {
"browser-ai-debate": {
"command": "cmd",
"args": ["/c", "node", "C:/path/to/browser-ai-debate-mcp/dist/index.js"],
"env": {
"BROWSER_AI_CDP_URL": "http://127.0.0.1:9222"
}
}
}
}For dedicated auto-launch mode, omit the env block.
Available Tools
1 toolbrowser_ai_debateA
Runs a structured multi-round debate between ChatGPT Web and Gemini Web using browser automation via CDP. Returns a structured conclusion with confidence score, rationale, and provider-level details. No API keys or external APIs are used.
| Name | Required | Description | Default |
|---|---|---|---|
| context | No | Additional context for the debate | |
| question | Yes | The question to debate between ChatGPT and Gemini | |
| finalizer | No | Which provider makes the final decision | |
| maxRounds | No | Maximum debate rounds (1-5) | |
| timeoutMs | No | Total timeout in milliseconds | |
| decisionCriteria | No | Decision criteria to evaluate conclusions against | |
| includeTranscript | No | Include full transcript in output | |
| providerTimeoutMs | No | Per-provider timeout in milliseconds |
Output Schema
| Name | Required | Description |
|---|---|---|
| audit | Yes | |
| rounds | Yes | |
| status | Yes | |
| maxRounds | Yes | |
| providers | Yes | |
| rationale | Yes | |
| agreements | Yes | |
| conclusion | Yes | |
| confidence | Yes | |
| durationMs | Yes | |
| nextActions | Yes | |
| totalRounds | Yes | |
| disagreements | Yes | |
| remainingRisks | Yes | |
| partialExplanation | No | |
| geminiConversationUrl | No | |
| chatgptConversationUrl | No |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description explains the method (browser automation via CDP) and output, but lacks details on potential side effects, reliability, or dynamic behavior. With no annotations, more transparency would be helpful, but the core behavior is conveyed.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three sentences, each serving a clear purpose: what it does, what it returns, and a key feature. No wasted words, highly efficient.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description covers input, process, and output. An output schema exists. It could mention error handling or limitations, but overall it's sufficiently complete for a tool with well-documented parameters.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the description does not need to add parameter meaning. It focuses on the overall tool purpose. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it runs a structured multi-round debate between ChatGPT Web and Gemini Web using browser automation via CDP, with a specific output. No siblings exist, so no need for differentiation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
While no explicit alternatives are given due to no siblings, the description highlights that no API keys or external APIs are used, which is a key usage hint. It could benefit from more explicit when-to-use guidance, but it's clear enough.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
1 tool update
v0.1.0- First observed
browser_ai_debate
TDQS
Only one tool exists, so there is no possibility of confusion or overlap with other tools.
With a single tool, naming consistency is not applicable; the name is descriptive and appropriate.
A single tool for a dedicated task is borderline; while it covers the stated purpose, a server with only one tool feels thin and limits flexibility.
The tool fully covers the intended debate workflow from start to conclusion. Minor gaps like configurable rounds are absent but not critical.
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
MCP server unifying ERPs, CRMs, APIs and knowledge base for Claude, ChatGPT and Gemini.
MCP server for AI dialogue using various LLM models via AceDataCloud
MCP server for building and testing AI agents with multi-model experimentation and insights.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceAn MCP server that enables multi-model debate and consensus building through a single tool. It orchestrates multiple AI models from various providers to debate topics and reach validated conclusions with real-time progress tracking.203MIT
- AlicenseNot gradedqualityCmaintenanceAn MCP server that enables multi-round AI expert debates with dynamic expert generation, cross-debate, and Tufte-style infographic export.6MIT
- AlicenseAqualityBmaintenanceMCP server that reduces confirmation bias in LLMs by orchestrating structured debates between asymmetric context sessions.13MIT
- AlicenseAqualityBmaintenanceDynamic multi-agent debate MCP server for structured argumentation and reasoning among AI agents.17MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/kaenozu/browser-ai-debate-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server