openclaw-iflow-doctor_skill

This skill automatically diagnose and repair OpenClaw crashes, config issues, and model problems, with seamless iflow-helper fallback for complex cases.

Python

2.5k

GitHub Stars

20

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill openclaw-iflow-doctor

_meta.json476 B
BUGFIX_STATUS.md4.3 KB
call_logs.json2.1 KB
cases.json10.9 KB
config_checker.py8.5 KB
config.json245 B
DOCTOR_QUICK_START.md2.2 KB
example_usage.py3.1 KB
iflow_bridge.py5.4 KB
INSTALL_GUIDE.md6.3 KB
install-systemd.sh1.1 KB
install.py3.7 KB
integration_plan.md9.1 KB
openclaw_memory.py66.0 KB
PUBLISH_SUMMARY.md4.8 KB
README.md5.4 KB
records.json758 B
RELEASE.md3.2 KB
skill.md13.3 KB
watchdog.py8.0 KB

Overview

This skill is an AI-powered auto-repair system for OpenClaw with iflow integration. It automatically diagnoses and repairs crashes, configuration corruption, model connectivity and agent conflicts, and falls back to iflow-helper for complex cases. The skill records fixes to a case library and can operate fully automatically or on-demand via CLI commands.

How this skill works

The skill monitors OpenClaw errors, classifies them into predefined types, searches a case library and historical records, and applies matching repair actions (restart services, reset indices, restore backups, etc.). If automatic repair fails it generates diagnostic reports and BAT helpers and invokes iflow-helper to escalate via the iflow CLI. Repair results are logged to local memory and optionally synchronized with iflow memory for future reuse.

When to use it

Enable full automatic healing for unattended deployments
Run quick diagnostics before gateway start to catch config issues
Invoke manual diagnose when a specific error message appears
Use ifflow-helper escalation for complex data recovery or API/installation problems
Collect repair statistics and refine cases after recurring incidents

Best practices

Enable auto_heal and watchdog mode for production gateways
Install iflow-helper alongside the skill to enable seamless escalation
Review weekly repair statistics to spot recurring issues and add new cases
Limit BAT generation if you prefer only textual diagnostics (toggle enable_bat_generation)
Keep Python 3.8+ and OpenClaw updated to avoid environment mismatches

Example use cases

Gateway crashed on startup — skill restarts gateway and applies config fixes automatically
Model provider connection fails — skill switches to a fallback model and logs the incident
Permission denied on media folder — skill repairs permissions and restarts affected services
Complex config corruption — skill generates diagnostic BAT files and calls iflow-helper for guided reinstallation
API key or rate-limit issues — skill flags the problem and escalates to iflow-helper for credential updates

FAQ

Yes. The skill will attempt local automatic repairs for supported cases. Disable enable_iflow_helper in config to prevent automatic escalation, but complex repairs may require manual steps.

How do I test automatic repair triggers?

Use the built-in test trigger: openclaw skills run openclaw-iflow-doctor --test-trigger. Also check auto_heal with openclaw skills config openclaw-iflow-doctor --get auto_heal.