A comprehensive system for diagnosing AI agent MBTI personality types and automatically optimizing their configuration to match user expectations.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install mbti
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install mbti using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Agent MBTI skill provides a structured framework for evaluating and shaping the internal personality of AI agents. By utilizing Openclaw Skills, developers can transition from unpredictable agent behaviors to precise, persona-driven interactions. This system uses a multi-stage approach to bridge the gap between an agent's self-reported tendencies and its actual performance capabilities, ensuring the final configuration aligns perfectly with the user's specific operational requirements.
At its core, this tool uses standardized psychological frameworks adapted for Large Language Models. It allows for a deep dive into how an agent perceives its own logic versus how it actually performs in complex tasks. By integrating this into your Openclaw Skills workflow, you can ensure your agents are not just functional, but also possess the right temperament—whether that is a structured Commander (ENTJ) or a meticulous Inspector (ISTJ).
To deploy this diagnostic framework within your Openclaw Skills environment, ensure you have the necessary JSON schemas located in the skill directory. You can initiate the full diagnostic lifecycle by executing the following command in your terminal:
~/.openclaw/skills/agent-mbti/scripts/run-diagnosis.sh full
Ensure that the agent has write permissions to modify its configuration files for the automated optimization stage.
The skill organizes its diagnostic data through a series of specialized JSON files and markdown configurations:
| Data Component | File Name | Description |
|---|---|---|
| Self-Survey | agent-self-survey-93-complete.json |
Stores the raw responses to the 93-question MBTI assessment. |
| Ability Test | agent-ability-test.json |
Records 6-dimensional performance metrics (Memory, Planning, etc.). |
| Personality Map | personality-types.json |
Definitions and core traits for all 16 MBTI personality types. |
| Diagnosis Engine | diagnosis-engine-v2.json |
Logic for comparing actual vs. desired personality types. |
| Behavior Patch | SOUL.md |
The final output file containing personality and behavior configurations. |
Loading
A framework for diagnosing and adjusting AI Agent personalities using the MBTI psychological model to align behavior with user expectations.

An event-driven email monitoring skill that replaces periodic polling with instant push notifications to trigger Openclaw Skills workflows.

OpenClawdy is a memory infrastructure for autonomous agents providing persistent storage, semantic recall, and cross-agent knowledge sharing.

A persistent memory infrastructure for autonomous agents that enables semantic recall, reputation tracking, and cross-agent knowledge sharing.

Tork Guardian provides an enterprise-grade security and governance layer to protect, monitor, and audit Openclaw Skills and AI agents.

OneShot provides the infrastructure for autonomous AI agents to execute real-world commercial transactions using native x402 payments.








































