Works with the AI tools you already use
Claude Code
Cursor
Codex CLI
GitHub Copilot
Gemini CLI+20 moreClaude CodeCursorCodex CLIGitHub CopilotGemini CLI
OpenClaw
Manus
Grok+17 agentsOpenClaw_doctor
Structured troubleshooting for OpenClaw skills, tools, and MCP connections using evidence-based diagnostics.
$9.99
OpenClaw_doctor
Example session with this skill installed
My OpenClaw agent can't see the new weather tool I added to the config. I've updated the .env file and restarted, but the logs show 'Skill discovery skipped: weather-tool'.
- Read your context and instructions
- Compiled the openclaw_doctor
- Generated the document
I have analyzed the skill discovery failure for your weather tool. The symptoms point to a configuration schema error or a path resolution issue within the OpenClaw environment.
OpenClaw Doctor Report
Symptom
Skill discovery failure (weather-tool skipped during startup).
Fault Domain
Skill Discovery / Configuration
Evidence
Log entry: Skill discovery skipped: weather-tool. User confirmed .env update and restart performed.
Diagnosis
| Status | Candidate | Evidence | Verification |
|---|---|---|---|
| LIKELY | Missing SKILL_PATH entry | Skill is skipped despite .env update | Check config.yaml or main entry point |
| CONFIRMED | Path resolution error | Log indicates discovery was attempted but bypassed | Verify absolute vs relative path in config |
| NEEDS VERIFICATION | Permission mismatch | None | Run ls -la on the tool directory |
Safe Fix
Update the config.yaml file to use the absolute path for the weather-tool directory instead of a relative path. OpenClaw discovery often fails when the execution context differs from the relative path root.
Rollback
Revert config.yaml to the previous relative path string: path: ./skills/weather-tool.
Verification
NOT RUN - Please update the path and provide the startup logs.
Security Notes
Ensure that the absolute path does not expose parent directories to the agent runtime. Do not change directory permissions to 777.
Next steps
- Locate the absolute path of the weather-tool folder.
- Update your OpenClaw configuration file with this path.
- Restart the agent and paste the new log output here.
openclaw-doctor.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
Debugging OpenClaw setups usually involves hours of digging through logs and guessing at environment variables. This skill provides a structured troubleshooting workflow to identify why your skills, tools, or MCP connections are failing.
What it does
- Symptom classification Categorizes issues into startup, execution, network, or permission domains for faster triaging.
- Evidence isolation Maps specific errors to the exact layer of the OpenClaw stack, from config to agent behavior.
- Root cause analysis Generates a diagnostic table documenting candidate causes, supporting evidence, and verification steps.
- Safe remediation Proposes minimal, reversible fixes with defined rollback procedures to prevent configuration drift.
- State verification Confirms fixes using narrow, safe checks rather than assuming success.
How it works
- Provide the error logs, configuration snippets, or a description of the failed behavior.
- The skill analyzes the supplied evidence to isolate the fault domain.
- You receive a diagnostic report including a candidate cause table and a specific fix.
- Apply the recommended fix and provide the new output for verification.
Frameworks & tools
This skill is designed specifically for the OpenClaw agent framework, including its MCP (Model Context Protocol) implementation, skill discovery system, and authentication layers.
Why this beats prompting it yourself
Generic troubleshooting prompts often suggest destructive "reinstall everything" actions or hallucinate configuration paths. This skill enforces a strict evidence-based workflow that prevents inventing system state and prioritizes security-conscious, reversible changes.
Use cases
- Debugging failed MCP tool connections or discovery errors.
- Resolving permission and environment variable conflicts during skill deployment.
- Fixing agent behavior issues where tools return data but the agent fails to process it.
- Auditing authentication flows and API key injection errors.
Known limitations
Cannot access your local filesystem or logs directly. You must provide the evidence manually. Does not support fixing underlying third-party API outages.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 5 days ago
- Passed all security checks, Safe to install