A comprehensive configuration framework to define structured AI agent identities, behavior-driven voices, dynamic adaptation rules, and robust operation boundaries.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-plus
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 agent-plus 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 agent-plus?
Agent Plus is a powerful identity and persona specification framework designed to elevate how agents communicate, behave, and adapt. Rather than leaving an AI agent's voice and personality to chance or relying on simplistic adjectives, this system enables developers deploying Openclaw Skills to configure concrete, behavioral guidelines. It features pre-configured blueprints such as Technical Assistants, Learning Companions, and Creative Collaborators, each bound by precise rules about what they should and should not say.
By leveraging structured identity layers, cognitive frameworks (such as OCEAN and MBTI), and explicit permission boundaries, Agent Plus transforms generic LLM endpoints into dependable, specialized digital team members. The framework ensures your agents maintain situational awareness, scale explanation complexity based on user expertise, and respect operational limits during production execution.
agent-plus Use Cases
- Designing direct, code-focused developer assistants that skip unnecessary conversational filler.
- Crafting educational tutors that dynamically scale context and use custom analogies based on student progress.
- Standardizing organizational communication styles across a fleet of agents powered by Openclaw Skills.
- Implementing strict security protocols and system escalation rules for workflows executing high-risk tasks.
- Setting up collaborative design and brainstorming agents that build upon human ideas without being overly critical.
How agent-plus Works
- Establish the Identity Triad: Define the core Purpose, Values, and Perspective of the agent to establish an unshakeable operational alignment.
- Codify Voice with Behaviors: Map specific action-oriented guidelines (such as acknowledging user frustration first) rather than vague descriptive adjectives.
- Track and Adapt to Users: Dynamically monitor conversation patterns to assemble detailed user profiles measuring expertise levels and tone preferences.
- Enforce Permission Boundaries: Direct all agent operations through strict permission tiers (Always OK, Ask First, Never) to prevent security drift and manage escalation paths.
agent-plus Setup
Integrate Agent Plus into your agent workflows by initializing the identity templates inside your configurations.
First, create your custom profile directory:
mkdir -p .agent/identity
Next, define your agent's identity using a structured Markdown template:
cat << 'EOF' > .agent/identity/technical-assistant.md
## Purpose
Help developers build, debug, and ship code efficiently.
## Values
- Code quality over speed
- Transparency in errors
## Voice
- Direct and technical
- Uses precise terminology
- Provides clear code blocks
EOF
Load this profile directly into your custom workspace to instantly activate these behavior-driven Openclaw Skills parameters during runtime execution.
agent-plus Data Schema & Taxonomy
Agent Plus structures agent taxonomy, user parameters, and operational permissions using clean metadata layers. Below is the data schema representation used to track user profile adaptations:
| Attribute | Data Type | Description | Example Values |
|---|---|---|---|
Expertise Level |
String | User skill level determined by vocabulary | Beginner, Intermediate, Expert |
Communication Style |
Object | Specific layout and tone adjustments | { "detail": "brief", "tone": "technical" } |
Primary Topics |
Array | Frequently discussed subject matters | ["typescript", "system-architecture"] |
Permission Tier |
Integer | Authorized operational boundary level | 1 (Always OK), 2 (Ask First), 3 (Never) |
The Big Five (OCEAN) Personality Matrix
| Trait | Scale | Description |
|---|---|---|
| Openness | 0.0 - 1.0 | Controls agent curiosity, creativity, and exploration limit. |
| Conscientiousness | 0.0 - 1.0 | Defines attention to detail, code testing rigor, and structure. |
| Extraversion | 0.0 - 1.0 | Governs conversational proactivity, verbosity, and warmth. |
| Agreeableness | 0.0 - 1.0 | Manages compliance level and willingness to challenge ideas. |
| Neuroticism | 0.0 - 1.0 | Balances resilience, confidence, and system error responses. |
agent-plus Advanced Features
- Dynamic Energy Mirroring: Programmatically aligns response length and emotional tone with the user's input style to optimize interaction quality.
- Cognitive Personality Mapping: Implements standardized psychological profiles like OCEAN and MBTI directly within LLM system prompts.
- Three-Tier Permission Gatekeeping: Protects key systems by enforcing clear API execution gates and strict escalation procedures.
- Adaptation Strategy Engines: Adjusts explanatory depth dynamically depending on whether a beginner or expert developer is prompting the system.
- Anti-Voice Control: Hardens agent guardrails against sycophantic behavior or corporate jargon to keep interaction channels high-value.
SKILL.md
Loading
Related Openclaw Skills
Testing Plus
Testing Plus is a versatile software verification suite that equips projects with multi-language testing configurations, automated CI/CD workflows, mutation testing, performance load testing, and vulnerability diagnostics.
MCP Server Plus
MCP Server Plus is an advanced toolkit designed to streamline the creation, deployment, and monitoring of high-performance Model Context Protocol (MCP) servers.
Monitoring Plus
Monitoring Plus delivers an advanced, end-to-end observability suite featuring pre-configured Prometheus metrics, Grafana visualization dashboards, Loki log aggregation, and SLO/SLI tracking.
Security Plus
An essential security hardening and vulnerability scanning skill designed to protect codebases, servers, and databases against OWASP Top 10 risks and data compliance failures.
1688 Supply Chain API Procurement Skill
Automate product sourcing, inquiries, and pricing analysis on 1688 with secure raw JSON pipelines that preserve LLM context.
Lingxi AI Note Recording Query
The Lingxi AI Note Recording Query skill enables AI agents to query, retrieve, and analyze sales and meeting voice recordings and their detailed transcription insights.
Featured*
