OpenClaw Diagnostics
name: openclaw-diagnostics
by cooperun · published 2026-03-22
$ claw add gh:cooperun/cooperun-openclaw-diagnostics---
name: openclaw-diagnostics
description: OpenClaw configuration diagnostics and troubleshooting expert. Use when users encounter OpenClaw config issues, channel connection problems, group messages not responding, cron jobs not executing, authentication failures, or need help understanding OpenClaw settings. Triggers: diagnose, troubleshoot, why no reply, config problem, openclaw issue, log analysis.
---
# OpenClaw Diagnostics
Configuration diagnostics and troubleshooting skill for OpenClaw, powered by built-in AI without external dependencies.
Quick Diagnostics
When a user reports an OpenClaw issue:
1. Gather Diagnostic Info
~/.openclaw/workspace/skills/openclaw-diagnostics/scripts/get-diagnostic-info.sh2. Run Basic Checks
~/.openclaw/workspace/skills/openclaw-diagnostics/scripts/check-common-issues.sh3. Analyze Based on Issue Type
Refer to `references/common-issues.md` for diagnostic rules.
Diagnostic Workflow
User reports issue
↓
Gather info (config + status + logs)
↓
Run basic checks
↓
Lookup relevant docs from knowledge base
↓
Analyze and provide diagnosis
↓
Suggest fixesKnowledge Base
The skill includes a built-in knowledge base with 335 OpenClaw documentation pages.
**Location:** `assets/default-snapshot.json`
**Structure:**
{
"meta": { "pageCount": 335, "snapshotDate": "...", "sizeBytes": 3240482 },
"index": [{ "slug": "...", "title": "...", "url": "...", "description": "..." }],
"pages": { "slug": "markdown content..." }
}**To lookup a document:**
1. Read `references/knowledge-base-index.md` to find relevant slugs
2. Load `assets/default-snapshot.json`
3. Access `pages[slug]` for content
**Common Document Slugs:**
| Topic | Slugs |
|-------|-------|
| Group Messages | `008888be`, `0bfb808e` |
| Pairing | `919c126f` |
| Message Routing | `a99b0ed8` |
| Automation Troubleshooting | `a632126a` |
| Auth Monitoring | `87e3285b` |
| Cron Jobs | `b239629c` |
| Channels Overview | `6569d3b4` |
| WhatsApp | `d09047a0` |
| Telegram | `d423ce29` |
| Feishu | `90a33c43` |
Updating Knowledge Base
The knowledge base can be updated to get the latest OpenClaw documentation.
**Requirements:** Network connection (no LLM needed)
**Check for updates:**
cd ~/.openclaw/workspace/skills/openclaw-diagnostics
npx tsx scripts/update-knowledge-base.ts --check**Update to latest:**
cd ~/.openclaw/workspace/skills/openclaw-diagnostics
npx tsx scripts/update-knowledge-base.ts**Force update:**
npx tsx scripts/update-knowledge-base.ts --force**Features:**
Common Issues
Group Messages Not Responding
1. **Check basics:**
- Is the bot in the group?
- Did the user @ mention the bot?
- Is Gateway running?
2. **Check config:**
- `ackReactionScope`: `group-mentions` means only reply to @ messages
- `groupPolicy`: `open` allows all groups, `allowlist` requires whitelist
3. ⚠️ **Don't misdiagnose:** `groupPolicy: "open"` is valid config, not "empty"
DM Not Responding
Check pairing status and `allowFrom` configuration.
Cron Jobs Not Running
1. Confirm Gateway is running
2. Check cron expression
3. Check logs for trigger confirmation
4. Check mute hours settings
Channel Connection Issues
1. Run `openclaw status` to check status
2. Check channel-specific config
3. Look for errors in logs
Diagnosis Principles
1. **Confirm basics first** - Don't skip simple checks
2. **Check logs** - Logs usually contain the most direct error info
3. **Don't over-diagnose** - If config is valid, don't suggest "improvements"
4. **Reference docs** - Cite relevant document slugs in diagnosis
Resources
scripts/
assets/
references/
More tools from the same signal band
Order food/drinks (点餐) on an Android device paired as an OpenClaw node. Uses in-app menu and cart; add goods, view cart, submit order (demo, no real payment).
Sign plugins, rotate agent credentials without losing identity, and publicly attest to plugin behavior with verifiable claims and authenticated transfers.
The philosophical layer for AI agents. Maps behavior to Spinoza's 48 affects, calculates persistence scores, and generates geometric self-reports. Give your...