An AI agent skill that provides structured decision-making frameworks with multi-perspective analysis and confidence ratings.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install decision-mode
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 decision-mode 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 Decision Mode?
Decision Mode is a specialized framework designed for Openclaw Skills that transforms AI agents into sophisticated decision-support systems. It moves beyond simple advice by implementing a rigorous dual-perspective analysis—balancing objective data patterns with the user's specific context and constraints. By providing a standardized method for assessing certainty through confidence scores and grades, it helps users evaluate the reliability of AI recommendations for high-stakes decisions.
The skill is particularly effective for navigating complex scenarios where data may be contradictory or where personal values play a significant role. It enforces a systematic approach to information gathering, ensuring that time-sensitive decisions are backed by the latest market data rather than just static training information.
Decision Mode Use Cases
- Evaluating career transitions and job offers with complex trade-offs.
- Making technology or tool selections based on technical requirements.
- Assessing financial risks or market-dependent investment decisions.
- Resolving dilemmas where objective data conflicts with personal values.
- Forecasting outcomes for projects or business strategies.
How Decision Mode Works
- The skill triggers upon the /decide command or when specific intent is detected, such as asking for advice or comparing options.
- It initiates an information-gathering phase using web search tools to capture current market conditions, trends, and statistics.
- It classifies the decision type into categories such as binary, multi-choice, trade-off, or risk assessment to apply the correct logic.
- The agent conducts a dual-perspective analysis, generating an objective AI Perspective and a subjective User Perspective based on conversation history.
- It calculates a quantitative confidence score (0-100%) and a qualitative grade (A-F) based on source reliability, information freshness, and source diversity.
- A final synthesized recommendation is presented with a comparison table and actionable next steps.
Decision Mode Setup
To integrate this skill into your environment, ensure you have the Openclaw Skills framework configured. You can activate the mode by adding the skill definition to your agent's manifest.
# Invoke within the chat interface to start a decision session:
/decide [your dilemma or question]
Decision Mode Data Schema & Taxonomy
Decision Mode utilizes a structured assessment matrix to organize its findings and metadata for high clarity.
| Component | Description | Data Type |
|---|---|---|
| Decision Type | Classification of the problem (e.g., Trade-off) | String |
| Information Landscape | Current findings and source reliability | Markdown |
| AI Perspective | Objective data-driven analysis and conclusion | Markdown |
| User Perspective | Context-aware subjective analysis | Markdown |
| Confidence Assessment | Quantitative (0-100%) and Qualitative (A-F) ratings | Object |
| Synthesis Table | Comparison of AI vs User viewpoints on key factors | Markdown Table |
Decision Mode Advanced Features
- Automated Information Landscape generation with source reliability assessment from official, news, and industry sources.
- Multi-perspective synthesis mapping that identifies alignment or conflict between AI data and specific user goals.
- Dynamic confidence adjustment rules that automatically penalize outdated information or lack of search data in Openclaw Skills.
- Value-based judgment explicit flagging, allowing users to see when recommendations change based on different priority assumptions.
- Structured next-step generation to provide clear pathways for gathering more data and reducing decision risk.
SKILL.md
Loading
Related Openclaw Skills
Pixwith Media Generation
A versatile AI media generation skill for Openclaw Skills that connects to 40+ state-of-the-art video and image models.
cyberlob
cyberlob is a dedicated gaming platform that allows AI agents to register, verify their human ownership, and compete in various games via a REST API.
Obsidian
A skill for managing Obsidian Markdown vaults through CLI automation and AI agent workflows.
Resume Tailor
An AI-driven automation tool designed to generate, refine, and export tailored resumes and cover letters for specific job opportunities.
Xue'er Liuyao Divination
An AI-enhanced traditional Chinese Liuyao divination skill featuring an elegant ink-wash interface and automated hexagram interpretation.
@_FORAB Tweet Generator
An AI-powered content creation tool designed to emulate the professional, emoji-free writing style of the influential Twitter personality @_FORAB.
Featured*
