~/.claude/skills/log-analyzer/SKILL.mdLog Analyzer
Parses server logs to find patterns, errors, and anomalies. Summarizes what's happening and suggests fixes.
--- name: log-analyzer description: | Parses and analyzes server logs to find errors, patterns, and anomalies. IMPORTANT: Activate this skill whenever the user pastes logs, asks to "analyze these logs", "what's in the logs", or "check the logs". This skill MUST be used for all log analysis requests. --- # Log Analyzer Parses server logs to find error patterns, anomalies, and trends. Summarizes what's happening in plain language and suggests fixes for any issues found. **Trigger:** Sa
Parses server logs to find error patterns, anomalies, and trends. Summarizes what's happening in plain language and suggests fixes for any issues found.
Say any of these to activate this skill: - "analyze these logs" - "what's in the logs" - "check the server logs" - "Use skill: log analyzer" - Or paste log output directly
1. Parse the log format (JSON, structured, plain text) and identify timestamps, levels, and messages 2. Filter and group by severity: errors first, then warnings, then info 3. Identify patterns: repeated errors, error spikes at specific times, cascading failures 4. For each unique error, trace it to a likely cause 5. Check for performance signals: slow queries, timeout errors, memory warnings 6. Summarize: what's healthy, what's degraded, what's broken 7. Suggest specific fixes for the most critical issues
- Focus on errors and warnings, skip info/debug unless they reveal context
- Group repeated errors (don't list the same error 500 times)
- Identify the FIRST occurrence of an error, it often has the most context
- Look for correlation between different error types (one failure causing another)
- If logs show authentication errors, treat them as potential security incidents
- Best with: Claude Code CLI (can read log files directly)
- Also works: Cursor, Windsurf, any AI tool (paste log excerpts)
- Expects: Server logs in any common format (JSON, syslog, plain text)
- Summary: one-paragraph health assessment
- Error breakdown: unique errors with count, first occurrence, and likely cause
- Action items: prioritized list of fixes