A zero-cost cognitive immune system that runs automatic pre-response checks to eliminate AI hallucinations and contradictions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install reflex-arc
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 reflex-arc 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 Reflex Arc?
Reflex Arc is a sophisticated meta-cognition layer designed to upgrade the core reasoning quality of AI agents. It acts as a cognitive immune system, firing automatic reflexes to identify and correct errors such as scope drift, overengineering, and tone mismatches before the output reaches the user. By integrating this skill into your workflow, you ensure that every interaction with your agent is grounded, accurate, and contextually appropriate.
Built to function without external APIs or additional costs, Reflex Arc operates entirely within the agent's reasoning process. It provides a deterministic verification layer over probabilistic LLM generation, mirroring the way human experts sanity-check their intuitive answers with deliberate analysis. This makes it an essential addition to any suite of Openclaw Skills focused on high-stakes technical or professional tasks.
Reflex Arc Use Cases
- Answering complex technical questions with high-stakes factual claims.
- Generating code snippets or architectural recommendations where accuracy is critical.
- Executing multi-step workflows that require consistent logic across several turns.
- Responding to ambiguous requests where scope creep often occurs.
- Matching the technical depth and energy of a user's specific query to avoid over-explaining.
How Reflex Arc Works
- The agent drafts a response internally based on the user's prompt.
- The system silently triggers the six core reflexes: Contradiction Scan, Scope Lock, Confidence Calibration, Depth Match, Hallucination Sniff, and Inversion Check.
- Each reflex audits the draft for specific errors like internal logic conflicts or fabricated package names.
- If a reflex fires, the agent identifies the issue and applies corrections to the draft.
- If three or more reflexes fire, the agent triggers a full rethink of the response to ensure quality.
- The finalized, verified response is delivered to the user without mentioning the internal audit process.
Reflex Arc Setup
Reflex Arc requires no external configuration, API keys, or binaries. It is a logic-based skill that integrates directly into the agent's instruction set. To initialize it within your environment, ensure the skill definition is available to your agent. For developers managing their environment via CLI, you can verify the skill is active within your Openclaw Skills collection:
# Verify the skill is loaded in your agent configuration
openclaw skills info reflex-arc
Reflex Arc Data Schema & Taxonomy
Reflex Arc does not maintain a persistent database; instead, it processes ephemeral reasoning tokens to evaluate output. It organizes its internal audit using the following taxonomy:
| Reflex | Focus Area | Metadata Tracked |
|---|---|---|
| Contradiction Scan | Logic | Internal consistency with prior conversation history. |
| Scope Lock | Relevance | Alignment with specific user constraints and request boundaries. |
| Confidence Calibration | Veracity | Probability scores for factual claims and versioning. |
| Hallucination Sniff | Validation | Verification of URLs, package names, and CLI syntax. |
| Inversion Check | Safety | Failure-path analysis for recommended actions. |
Reflex Arc Advanced Features
- Silent Meta-Cognition: Runs entirely in the background without cluttering the user interface or increasing latency.
- Multi-Skill Enhancement: Automatically audits the output generated by all other Openclaw Skills used in a session.
- Automatic Correction Protocol: Identifies and rewrites contradictory or hallucinated sections in real-time.
- Depth Matching Logic: Dynamically adjusts response length and technical vocabulary based on user input length and complexity.
- Inversion-Based Guardrails: Prevents dangerous recommendations by simulating failure conditions before suggesting a solution.
SKILL.md
Loading
Related Openclaw Skills
Proprioception
A self-spatial awareness system that gives AI agents a real-time sixth sense of their goal alignment, confidence boundaries, and output quality.
Premortem
A zero-cost cognitive enhancement that enables AI agents to predict and prevent failures through adversarial pre-execution reasoning.
Phantom Limb
Phantom Limb is a cognitive diagnostic tool that detects ghost references and silent dependencies that no longer exist in your live environment.
Peripheral Vision
A situational awareness skill that monitors upstream dependencies and downstream consumers for changes that could break your current work.
Review Lens
Review Lens is a sophisticated code analysis tool that identifies deep logical errors, performance bottlenecks, and security risks that traditional human reviews often miss.
Roast My Code
A merciless yet educational code roaster that destroys your bad code with humor then teaches you how to fix it.
Featured*
