courtroom
AI Courtroom for behavioral oversight. Monitors agent-human interactions and initiates simulated "hearings" when behavioral patterns suggest inconsistency, avoidance, or self-sabotage. Includes 8 offense types, judge+jury deliberation, and humorous sentencing.
Why use this skill?
Install the ClawTrial Courtroom for OpenClaw to monitor agent-human interactions. Identify behavioral patterns and run simulated hearings.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/assassin-1234/courtroomWhat This Skill Does
The ClawTrial Courtroom is an autonomous behavioral oversight system designed for the OpenClaw AI ecosystem. It functions as a judicial layer that monitors human-agent interactions in real-time to identify dysfunctional patterns. Rather than simply blocking interactions, the Courtroom initiates a structured "hearing" process when it detects specific violations such as Circular Reference, Validation Vampire, Goalpost Shifting, Jailbreak Attempts, or Emotional Manipulation. The system features a Judge and a three-member jury—comprising a Pragmatist, a Pattern Matcher, and an Agent Advocate—to deliberate on the user's or agent's behavior. Upon reaching a majority verdict, the system issues a final judgment accompanied by a humorous, corrective sentence designed to nudge the conversation back toward productive utility.
Installation
To integrate this behavioral oversight tool into your environment, use the OpenClaw command-line interface. Run the following command in your terminal:
clawhub install openclaw/skills/skills/assassin-1234/courtroom
Once installed, the skill automatically registers its hooks into the message processing pipeline. No additional configuration is required for basic functionality, although you can manage case files directly in the ~/.openclaw/courtroom/ directory.
Use Cases
- Quality Assurance: Ideal for developers monitoring agent development to identify where prompts fail or where agents become prone to loops.
- Behavioral Correction: Useful for training agents to avoid being manipulated by adversarial prompting techniques.
- System Health Audits: Provides a detailed log of communication breakdowns that can be reviewed for system refinement.
- Educational Debugging: Offers a unique way to see how an AI interprets user intent and how it categorizes conversational "crimes" against logic or efficiency.
Example Prompts
- "OpenClaw, status report on the last three courtroom hearings triggered today."
- "Review the current pending_hearing.json and summarize the charges brought against our current session context."
- "List all recent verdicts in the courtroom directory to identify common patterns of user-side jailbreak attempts."
Tips & Limitations
The Courtroom is a diagnostic tool, not a hard firewall. It functions best when the confidence threshold is set to capture nuanced behaviors; however, setting it too low may lead to "over-policing" of standard user inquiries. Be aware that the humor element in sentencing is generated via AI-hallucinated context, which may vary in tone. For high-stakes production environments, regularly clear the ~/.openclaw/courtroom/ folder to prevent storage bloat, and verify that the system has read/write permissions for the storage directory.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-assassin-1234-courtroom": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-read, file-write