Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
FACEPALM reads OpenClaw console logs from the last 5 minutes alongside the active session's chat history, then sends that combined context to Codex 5.3 for diagnosis. It runs manually on demand or fires automatically when Agent Swarm detects a troubleshooting loop.
Cross-referencing the error log with what the user was actually doing at the time produces a diagnosis grounded in session context, not just the raw stack trace.
When to use it
- Diagnosing why a gateway error keeps recurring despite repeated fixes
- Correlating a console error with what you typed or triggered minutes earlier
- Getting actionable remediation steps after a failed skill invocation
- Automatically breaking out of agent troubleshooting loops
- Reviewing session context alongside raw logs without doing it by hand
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: why does gateway keep failing on my last request?
- 1Reads gateway.log entries from the past 5 minutes
- 2Extracts chat history from the active session transcript
- 3Crosschecks console errors against the chat context to find correlations
- 4Sends combined data to Codex 5.3 via OpenRouter for analysis
- 5Returns a structured diagnosis with specific remediation steps
Root cause identified (e.g. malformed request payload from a prior skill call) with exact steps to resolve it
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.