Chain Reason is an explicit reasoning trace skill that provides auditable, multi-step logical workflows for complex decision-making and problem-solving.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install chain-reason
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 chain-reason 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 Chain Reason?
Chain Reason is a specialized logical framework designed to bring transparency and auditability to AI decision-making. By implementing rigorous reasoning traces, it ensures that every step of a complex operation is visible, labeled, and verifiable, adhering to the high standards of Openclaw Skills and arifOS constitutional floors.
This skill forces the AI to 'show its work,' moving beyond simple answers to provide a structured breakdown of givens, constraints, and intermediate steps. It is particularly effective for systems where human authorization and technical stability are paramount, ensuring that no assumption goes unchecked and no constraint is silently bypassed.
Chain Reason Use Cases
- Resolving multi-constraint problems where requirements may conflict.
- Providing transparency when a user asks to see the logic or explanation behind a decision.
- Weighing tradeoffs between multiple valid architectural or coding approaches.
- Conducting safety audits before high-stakes actions like production deployments.
- Decomposing complex queries that require more than three distinct logical leaps.
How Chain Reason Works
- Activation Trigger: The skill engages when it detects keywords like 'logic', 'why', or 'reason', or when a task involves high complexity.
- Givens and Constraints Identification: The agent first establishes what facts are known and what boundaries must be respected.
- Approach Mapping: A strategy is defined to navigate from the current state to the desired solution.
- Iterative Step Processing: Each logical step is processed, numbered, and recorded to maintain a clear audit trail.
- Verification and Verdict: The final conclusion is checked against constraints and assigned a status such as SEAL, CAUTION, or HOLD.
Chain Reason Setup
To integrate Chain Reason into your arifOS environment, include the skill definition in your agent's capability manifest. Ensure the Openclaw Skills metadata is properly configured:
# Example of referencing the skill in your configuration
openclaw install chain-reason
Ensure that your agent is configured to recognize the CHAIN REASONING, TRADEOFF ANALYSIS, and EVIDENCE CHAIN block formats in its output templates.
Chain Reason Data Schema & Taxonomy
The skill produces structured markdown outputs categorized by the type of reasoning required:
| Trace Type | Key Components | Purpose |
|---|---|---|
| CHAIN REASONING | GIVEN, CONSTRAINT, APPROACH, STEP-N, CONCLUSION | Standard logical flow and constraint checking. |
| TRADEOFF ANALYSIS | [+W] Factors, [-W] Factors, NET, RISK, MITIGATION | Deciding between competing options with weighted risks. |
| EVIDENCE CHAIN | OBS, DER, INT, SPEC, CONFIDENCE, GAP | Knowledge-based questions requiring factual derivation. |
Chain Reason Advanced Features
- Multi-Agent Audit Support: Reasoning traces can be handed off to secondary 'auditor' agents for verification.
- Constitutional Floor Alignment: Automatically triggers a HOLD status if actions violate pre-defined safety floors (e.g., Floor 13 for Kedaulatan).
- Epistemic Labeling: Distinguishes between direct observations and interpretations to manage uncertainty.
- Tradeoff Weighting: A specialized sub-workflow for balancing competing technical or business interests.
- Confidence Scoring: Provides a high/medium/low confidence rating based on the missing evidence identified in the GAP section.
SKILL.md
Loading
Related Openclaw Skills
Support Flow Builder
A documentation-first tool to convert support policies and ticket patterns into structured service workflows and agent playbooks.
Repository Architecture Visualizer
Automatically transform any local code repository into an interactive, hierarchical architecture visualization rendered as a portable HTML page.
Payment Route Optimizer
A strategic planning tool for building payment-routing diagnostics, failover policies, and merchant rollout briefs.
SVN Tool
A powerful Subversion version control skill for managing repository operations, branching, and conflict resolution directly through CLI integration.
Algorithm Mentor
A Socratic teaching agent that guides developers through algorithm challenges without giving away answers immediately.
Commerce BI Copilot
An intelligent ecommerce analysis tool that transforms raw data notes and KPI exports into structured operator-ready briefs and prioritized action plans.
Featured*
