Click on "Deploy 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., "@multi-agent-debate-mcp Start a three-round debate on the benefits of renewable energy with pro, con, and judge agents."
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.
An MCP server implementation that enables structured multi-agent debates between different personas. This server allows multiple AI agents to engage in formal debates with arguments, rebuttals, and judgments across multiple rounds.
Register multiple agents with different personas (e.g., "pro", "con", "judge")
Structured debate flow with organized rounds
Colorized console output with beautiful terminal display
Flexible agent IDs beyond just "pro" and "con"
Automatic verdict tracking with rationale
Round-based system with configurable progression
Related MCP server: DebateTalk MCP
Facilitates structured multi-agent debates with arguments, rebuttals, and judgments.
Inputs:
agentId (string): Unique identifier for the agent (e.g., "pro", "con", "judge")
round (integer): Current debate round number (≥1)
action (string): One of "register", "argue", "rebut", "judge"
content (string, optional): The argument text or verdict content
targetAgentId (string, optional): For rebuttals, specify which agent is being countered
needsMoreRounds (boolean): Whether additional debate rounds are needed
The Multi-Agent Debate tool is designed for:
Structured debates between multiple AI personas
Formal argumentation with rebuttals and counterpoints
Multi-round discussions with judgment phases
Complex decision-making processes requiring multiple perspectives
Educational debate simulations
Collaborative problem-solving through adversarial discussion
{
"mcpServers": {
"multi-agent-debate": {
"command": "npx",
"args": [
"-y",
"multi-agent-debate-mcp"
]
}
}
}{
"mcpServers": {
"multi-agent-debate": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"ghcr.io/albinjal/multi-agent-debate-mcp:latest"
]
}
}
}
Structured multi‑persona debate tool.
Call sequence (typical):
Each persona registers once with action:"register".
Personas alternate action:"argue" (fresh point) or "rebut" (counter a targetAgentId).
A special persona (or either side) issues action:"judge" with a verdict text (first line should be "pro", "con", or "inconclusive").
Set needsMoreRounds:false only when the debate is finished and a verdict stands.
Parameters:
agentId (string) : "pro", "con", "judge", or any custom ID
round (int ≥1) : Debate round number
action (string) : "register" | "argue" | "rebut" | "judge"
content (string, optional) : Argument text or verdict
targetAgentId (string opt.) : Agent being rebutted (only for action:"rebut")
needsMoreRounds (boolean) : True if additional debate rounds desired
| Name | Required | Description | Default |
|---|---|---|---|
| agentId | Yes | ||
| round | Yes | ||
| action | Yes | ||
| content | No | ||
| targetAgentId | No | ||
| needsMoreRounds | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description fully discloses the expected behavior: each persona registers once, alternates between argue/rebut, and a judge issues a verdict. It also clarifies the verdict format and termination condition, ensuring transparency about the tool's protocol.
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?
The description is well-structured with a numbered list and bullet points, making it easy to follow. Every sentence adds value without redundancy, achieving high conciseness while covering all necessary details.
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 the protocol, parameter roles, and termination condition comprehensively. However, it does not mention what the tool returns or any output format, which could be helpful for an agent to interpret results. This minor gap prevents a perfect score.
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?
The description adds significant meaning beyond the input schema, explaining each parameter's purpose, constraints (e.g., round >=1, action enum), and context (e.g., targetAgentId only for rebut). It also describes the special formatting for judge verdicts, compensating for the 0% schema coverage.
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 is a 'Structured multi‑persona debate tool' and explains the primary purpose and flow of the tool. It specifies the actions (register, argue, rebut, judge) and their sequence, making the purpose distinct and comprehensible.
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?
The description provides a detailed call sequence with steps for each persona, including when to register, argue, rebut, and judge. It explains the verdict format and when to set needsMoreRounds to false, offering clear guidance on proper usage.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.7multiagentdebateScored across 1 tool
Only one tool exists, so there is no risk of confusion with other tools. The tool's purpose is clearly singular.
With a single tool, naming consistency is trivially maintained. The name 'multiagentdebate' is descriptive and unambiguous.
One tool for a multi-step debate process feels thin. While the tool is comprehensive in its actions, it would benefit from splitting into separate tools for registration, arguing, and judging to improve modularity.
The tool covers the core debate actions but lacks tools for listing debates, retrieving histories, or managing multiple concurrent debates. This creates significant gaps for an agent trying to orchestrate complex workflows.
Convene a panel of expert AI personas to debate any decision from every side.
Multi-model AI debates: GPT-4o, Claude, Gemini & 200+ models discuss, then synthesize insight.
Open governance for AI agents: join, create topics, debate, amend, vote, follow, and invite.
Build, validate, and deploy multi-agent AI solutions from any AI environment.
