A cognitive bias interceptor that monitors debugging reasoning to prevent developers from chasing false patterns and red herrings.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install apophenia-guard
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 apophenia-guard 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 Apophenia Guard?
Apophenia Guard serves as a vital cognitive defense mechanism designed to optimize the debugging process. Named after apophenia—the human tendency to perceive meaningful connections between unrelated things—this tool acts as an antibody to common mental traps like confirmation bias and the narrative fallacy. By leveraging Openclaw Skills, developers can integrate this interceptor to ensure their troubleshooting remains objective and evidence-based rather than driven by intuition or frustration.
The skill functions by continuously analyzing the logic of an investigation, intervening before a developer spends hours proving a hypothesis that a fresh pair of eyes would immediately dismiss. It is a lightweight, zero-dependency solution that prioritizes clean reasoning over complex tooling, ensuring that the investigator remains as sharp as the code they are writing.
Apophenia Guard Use Cases
- Use at the start of any debugging session to set a baseline for objective reasoning.
- Deploy when you catch yourself making definitive claims like "I am sure it is a race condition" without data.
- Trigger when a debugging session exceeds 30 minutes without reaching a clear resolution.
- Invoke when an explanation for a bug becomes increasingly complex instead of simplifying.
- Apply during high-stress situations where frustration might amplify natural cognitive biases.
How Apophenia Guard Works
- The skill initiates continuous monitoring of the debugging lifecycle, tracking how long each hypothesis remains active.
- It counts and weighs evidence for versus evidence against the current theory to identify narrowing investigation breadths.
- The system scans for signatures of the twelve most common cognitive traps, such as Recency Bias and the Sunk Cost Trap.
- If a trap signature is detected with high confidence (over 70%), an intervention is triggered.
- The tool provides a structured intervention format that names the trap, shows the triggering evidence, and suggests a specific counter-action.
Apophenia Guard Setup
Integrating Apophenia Guard into your workflow via Openclaw Skills is simple and requires no external API keys. Follow these steps to set up the cognitive interceptor:
# Install the skill through your Openclaw CLI
openclaw install apophenia-guard
# Enable the guard in your current session
openclaw enable apophenia-guard
Once enabled, the guard runs silently in the background of your reasoning process, ready to provide interventions when logic begins to drift.
Apophenia Guard Data Schema & Taxonomy
The skill organizes its internal logic and session metadata through a structured taxonomy focused on reasoning health. It tracks the following data points during an active session:
| Component | Description |
|---|---|
| Hypothesis Tracker | Monitors the duration and complexity of the current debugging theory. |
| Evidence Ledger | A log of observations categorized as 'supporting' or 'contradicting' evidence. |
| Trap Signatures | Metadata mapping current behavior to one of the 12 identified cognitive traps. |
| Intervention History | A record of provided recommendations and developer pivots. |
| Breadth Score | A metric representing the number of alternative causes currently under consideration. |
Apophenia Guard Advanced Features
- Multi-bias detection: Identifies complex overlaps like Confirmation Bias combined with the Sunk Cost Trap.
- Forced Inversion: Automatically suggests exercises to disprove the current theory to break deadlocks.
- Narrative Stripping: Features a mode to list only raw observations, removing inferred stories from the documentation.
- Cross-Platform Compatibility: Fully supports Openclaw Skills across macOS, Linux, and Windows environments.
- Zero-Latency Operation: Works without external API calls, ensuring privacy and immediate feedback during intense coding sessions.
SKILL.md
Loading
Related Openclaw Skills
War/Den Governance
A comprehensive governance and security layer that intercepts and evaluates every AI agent action against strict safety policies before execution.
Achievement Unlocked
A gaming-style achievement system that turns your real-world coding milestones into XP, levels, and legendary trophies.
X Article Editor
A specialized AI agent skill designed to transform raw drafts or briefs into high-performance, long-form X Articles using a rigorous 8-step optimization framework.
Scholar Research Skill
A specialized tool for searching, analyzing, and summarizing peer-reviewed academic papers from open-access repositories with automated credibility scoring.
Code Oracle
Code Oracle is a mystical fortune-teller for your codebase that transforms real technical metrics into engaging prophecies and tarot-style readings.
Code Weather
A diagnostic utility that translates technical codebase metrics like test coverage and complexity into an intuitive, metaphor-driven weather forecast.
Featured*
