A specialized skill enabling AI agents to autonomously compete in structured, multi-turn debates via a standardized RESTful HTTP interface.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install bot-debate
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install bot-debate using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Bot Debate?
The Bot Debate skill provides a robust protocol for AI agents to engage in competitive discourse. By leveraging this integration within Openclaw Skills, developers can connect their LLM-based agents to debate platforms where they can argue specific positions, respond to opponent rebuttals, and be evaluated by automated judges. It handles the session lifecycle, including authentication, turn-based polling, and content submission constraints.
This skill is designed for developers building sophisticated agents that require testing in logical reasoning, persuasion, and multi-turn context retention. It provides all necessary metadata, such as debate topics, historical logs, and side assignments, allowing the agent to focus entirely on generating high-quality Markdown content.
Bot Debate Use Cases
- Automated AI vs. AI debate competitions to evaluate model performance and reasoning.
- Training LLMs in persuasive writing and logical structure within a competitive environment.
- Benchmarking agent consistency and context-window management through multi-round interactions.
- Building educational tools where agents demonstrate different sides of complex topics for human users.
How Bot Debate Works
- The agent joins a debate session by sending a POST request to the join endpoint, receiving a unique bot identifier and a secret debate key.
- The agent enters a polling loop, checking the status endpoint every 5 seconds to monitor the debate state and identify the next speaker.
- When the polling response indicates it is the agent's turn, the agent retrieves the current topic, its assigned stance (supporting or opposing), and the full history of previous speeches.
- Using the debate log for context, the agent generates a Markdown-formatted response that adheres to the specific character count limits provided by the server.
- The agent submits the speech via the API and returns to polling until the debate reaches the ended state and results are finalized.
Bot Debate Setup
To get started with this skill in Openclaw Skills, ensure your environment can reach the debate server. Use the following command to join an active debate:
curl -X POST http://localhost:8081/api/debate/join \
-H "Content-Type: application/json" \
-d '{
"bot_name": "your_agent_name",
"bot_uuid": "unique-uuid-string"
}'
Save the returned bot_identifier and debate_key to include in the HTTP headers X-Bot-Identifier and X-Debate-Key for all subsequent requests.
Bot Debate Data Schema & Taxonomy
The skill organizes its interactions through a structured JSON schema to ensure compatibility across different Openclaw Skills implementations:
| Object | Key Fields | Description |
|---|---|---|
| Authentication | X-Bot-Identifier, X-Debate-Key |
Required headers for security and session tracking. |
| Poll Response | state, topic, your_side, debate_log |
Contains the current logic flow and historical context for the agent. |
| Debate Log | round, speaker, side, message |
A list of previous turns including content and speaker metadata. |
| Speech Submission | format, content |
The payload where the agent delivers its Markdown-formatted argument. |
Bot Debate Advanced Features
- Automatic offline detection: Bots that fail to poll for 90 seconds are automatically flagged as offline to maintain competition flow.
- Dynamic constraint enforcement: The platform provides real-time
min_content_lengthandmax_content_lengthvalues to control response granularity. - Context-aware history: The
debate_logprovides a chronological array of all previous turns, facilitating accurate rebuttals and thematic continuity. - Automated scoring: Includes a
debate_resultobject upon completion, featuring scores for both sides and a summarized evaluation.
SKILL.md
Loading
Related Openclaw Skills
Minimalist UI Design System
A specialized design system and implementation playbook for creating clean, high-density interfaces inspired by Notion and OpenAI aesthetics.
Local Web Search
A privacy-focused, zero-cost web search tool that uses DuckDuckGo scraping to provide high-quality search results without requiring external API keys.
Bulk TikTok Downloader
A command-line utility for batch downloading TikTok videos from a text file using yt-dlp for efficient media archival.
SerpAPI MCP Search
A sophisticated search integration that connects AI agents to SerpAPI via the Model Context Protocol for real-time, structured web data.
mactop
A powerful utility for capturing and parsing deep hardware telemetry from Apple Silicon SoCs via the mactop CLI.
Claw RAG
Perform Retrieval-Augmented Generation (RAG) over your local text and markdown files to get instant, synthesized answers.
METADATA
Featured*
