How are the findings categorized?

Findings are classified into four levels: CRITICAL (requires immediate fix), WARNING (potential issues), INFO (recommendations), and PASS (optimal status).

What parts of the system are audited?

The audit covers 10 domains: Gateway, Security, Channels, Auth & Models, Memory, Context/Compaction, Heartbeat/Cron, Hooks & Skills, Workspace Integrity, and System/Network.

When should I use the openclaw-doctor skill?

You should use it whenever your agent is behaving unexpectedly, experiencing high latency, failing to access memory, or when you need to verify the security of your configuration files.

Does OpenClaw Doctor automatically fix issues?

No, OpenClaw Doctor is strictly read-only. It identifies problems and provides a recommended Quick Fix script for you to review and execute manually.

OpenClaw Doctor

Name: OpenClaw Doctor
Author: AnthemFlynn

byAnthemFlynn

•

分析与监控

Performs a comprehensive 10-domain diagnostic audit of OpenClaw agent infrastructure to identify health issues and recommend actionable fixes.

OpenClaw Doctor is a specialized diagnostic tool designed for 'agent-on-agent' infrastructure health checks. It systematically audits ten critical domains of an OpenClaw environment—including gateway status, security permissions, memory systems, and workspace integrity—to ensure peak operational performance. By classifying findings into severity levels (Critical, Warning, Info, Pass), it provides developers with a structured health report and a ready-to-use remediation script. This skill is ideal for troubleshooting connectivity issues, verifying security hardening, and optimizing context pruning without the risk of accidental system mutations.

主要功能

013 GitHub stars

02Read-only safety protocol that recommends fixes without modifying system state.

03Workspace integrity verification for essential agent files and identity documents.

04Comprehensive 10-domain audit covering gateway, security, memory, and networking.

05Automated 'Quick Fix' script generation for rapid manual remediation.

06Severity-based reporting to prioritize critical infrastructure failures.

使用场景

01Troubleshooting gateway connection timeouts or high-latency responses.

02Optimizing memory compaction and context pruning for improved AI agent efficiency.

03Auditing file permissions and security configurations for API key protection.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add anthemflynn/dwc openclaw-doctor

For use in Claude.ai and ChatGPT