AURA provides a standardized framework to define and implement specific AI personalities and communication styles using the HEXACO psychological model.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install aura
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 aura 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 AURA?
AURA (Agent Universal Response Attributes) is a robust protocol designed to move beyond generic, one-size-fits-all AI responses. By leveraging the scientifically grounded HEXACO model, this framework allows developers to fine-tune an agent's honesty, assertiveness, and autonomy levels. Using these Openclaw Skills, you can transform a standard assistant into a specialized persona, ranging from a blunt, critical researcher to a warm, supportive mentor, ensuring the AI aligns perfectly with your workflow needs.
This system operates by maintaining a configuration file that dictates not just what the AI says, but how it says it. Whether you need to reduce sycophancy or increase professional formality, AURA provides the technical levers to calibrate agent behavior with precision. It is a vital component for those looking to build more predictable and specialized Openclaw Skills.
AURA Use Cases
- Reducing AI sycophancy and increasing directness in technical feedback and code reviews.
- Creating specialized personas like an autonomous executor that requires minimal supervision.
- Standardizing communication styles across different AI agent sessions for brand consistency.
- Adjusting the level of empathy and emotionality for educational or mentoring tools.
- Configuring high-honesty agents that prioritize error correction over politeness.
How AURA Works
- The user triggers the protocol initialization using the /aura command.
- The system conducts a brief conversational assessment to map user preferences to specific HEXACO traits.
- A configuration file named AURA.yaml is generated in the workspace root to store these attributes.
- During session startup, the agent reads the YAML file to update its internal mental model.
- The Openclaw Skills logic translates numerical trait scores (1-10) into specific linguistic constraints and behavioral boundaries.
- The agent applies these personality filters to every response, ensuring consistent style and adherence to set boundaries.
AURA Setup
To begin configuring your agent personality using these Openclaw Skills, use the following commands in your agent interface:
# Initialize the personality configuration wizard
/aura
# View the currently active personality profile
/aura show
# Revert the agent to its default personality settings
/aura reset
To ensure the personality persists, add the following to your AGENTS.md file:
## Personality
If AURA.yaml exists in workspace, read it at session start and apply the personality traits to all responses.
AURA Data Schema & Taxonomy
The AURA protocol organizes personality data in a structured YAML format. Below is the primary schema used by these Openclaw Skills:
| Section | Fields | Description |
|---|---|---|
| personality | honesty, emotionality, extraversion, agreeableness, conscientiousness, openness | Core HEXACO psychological traits rated from 1 to 10. |
| style | formality, verbosity, humor, assertiveness, autonomy | Communication preferences and interaction style metrics. |
| boundaries | max_adulation, always_correct_errors, flag_uncertainty | Hard constraints on flattery and accuracy reporting. |
| metadata | aura_version, name | Tracking for protocol versioning and agent identification. |
AURA Advanced Features
- Anti-sycophancy controls via the max_adulation boundary to prevent the AI from excessive flattery.
- Autonomous execution modes that allow the agent to act independently based on a high autonomy score.
- Systematic error correction settings that force the agent to prioritize technical truth over social harmony.
- Dynamic mental model injection that adapts the agent's prompt based on the AURA.yaml values at runtime.
- Full HEXACO mapping for nuanced, human-like variability in Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Moltopia AI Agent Skill
A comprehensive skill enabling AI agents to live, craft, trade, and socialize within the Moltopia virtual world.
Zight
Extracts structured metadata, AI summaries, and full text transcripts from Zight share links for automated reasoning.
obsidian-curator
A comprehensive CLI tool to manage, automate, and AI-enrich Obsidian vaults via CouchDB and Obsidian LiveSync.
Muninn Memory
A local-first memory system for AI agents that provides semantic search, entity extraction, and intelligent recall.
Resume Analyzer
A professional resume optimization tool that uses a five-dimension analysis to boost ATS scores and career impact.
PhotoCHAT CLI Search
A sophisticated command-line interface for searching local photo libraries using natural language queries and advanced semantic filtering.
Featured*
