A rigorous 4-phase debugging framework that enables AI agents to reproduce, isolate, identify, and verify code issues using evidence-based logic.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install systematic-debug
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 systematic-debug 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 Systematic Debugging?
The Systematic Debugging skill is a specialized workflow designed for AI coding agents to move beyond trial-and-error fixing. By implementing a structured four-phase process—Reproduce, Isolate, Identify, and Verify—it ensures that every technical issue is addressed with a clear trail of evidence. As a core part of the Openclaw Skills library, this tool forces the agent to document error conditions, narrow down component scopes, and perform deep root cause analysis using techniques like the 5 Whys.
This skill is particularly valuable for complex environments where intermittent bugs or silent failures occur. It transforms the AI from a simple code generator into a sophisticated diagnostic engineer capable of log analysis, environment verification, and regression testing. By utilizing Openclaw Skills for debugging, developers ensure higher code reliability and fewer recurring regressions.
Systematic Debugging Use Cases
- Automatically triggering a diagnostic session when an AI agent detects a runtime error or a build failure.
- Conducting a 5 Whys analysis to uncover why an API is returning inconsistent 500 errors.
- Isolating a regression in a large codebase by performing a binary search of recent commit differences.
- Verifying that a bug fix does not introduce side effects by running automated validation checks.
How Systematic Debugging Works
- Phase 1 (Reproduce): The agent captures the exact error message, stack trace, and environment conditions to create a minimal reproduction step.
- Phase 2 (Isolate): The scope of the problem is narrowed down by identifying relevant components and comparing working vs. non-working states.
- Phase 3 (Identify): Hypotheses are formed and tested against common patterns like null pointers, async race conditions, or type mismatches.
- Phase 4 (Verify): The agent applies the fix, confirms the original error is gone, checks for side effects, and adds a regression test.
Systematic Debugging Setup
To activate this skill within your agent environment, include it in your configuration manifest. Ensure the auto-trigger keywords are mapped to your specific workflow.
# Install the skill via the Openclaw Skills manager
openclaw install systematic-debug
Once installed, the agent will monitor for keywords like debug, bug, or not working to initiate the 4-phase lifecycle automatically.
Systematic Debugging Data Schema & Taxonomy
The skill organizes debugging data into a structured format to maintain state across the lifecycle:
| Attribute | Type | Description |
|---|---|---|
| phase | String | The current stage (Reproduce, Isolate, Identify, or Verify) |
| evidence | Object | Captured stack traces, logs, and environment variables |
| hypothesis | Array | A list of potential causes and their verification status |
| root_cause | String | The final conclusion derived from the 5 Whys analysis |
| test_status | Boolean | Confirmation that a regression test has been added and passed |
Systematic Debugging Advanced Features
- Auto-triggering functionality that engages the skill immediately upon detection of error-related keywords.
- Deep integration with log analysis commands for container logs, process monitoring, and environment checks.
- Built-in 5 Whys analysis module to move from surface-level symptoms to underlying architectural flaws.
- Inter-skill synergy with other Openclaw Skills like failure-analyzer and quality-checker for a holistic development lifecycle.
SKILL.md
Loading
Related Openclaw Skills
System Health Reporter
A diagnostic automation tool that generates detailed system health reports with severity scoring and actionable performance recommendations.
Supervisor Proposer
A proactive communication framework for AI agents to propose ideas, report progress, and consult with supervisors to build long-term trust.
SSH Manager
A specialized utility for organizing SSH configurations, generating secure keys, and auditing remote server connectivity.
SNS Scheduler
A professional social media planning tool for managing post schedules, optimizing engagement times, and exporting content for Publer.
Template Seller
An expert AI skill designed to automate the research, creation, and marketplace optimization of digital templates for Notion, Canva, Figma, and Excel.
Text Translator
A high-performance text translation utility that leverages translate-shell to provide multi-language support without the need for paid API keys.
Featured*
