Skill: devops-monitor

Version: 1.0.0

Automated DevOps monitoring with AI-powered analysis.

Monitors logs, metrics, and codebase for issues.

Runs while you sleep - alerts on critical issues only.

Features:

- Log analysis and anomaly detection

- GitHub workflow monitoring

- Server health checks

- Automated debugging suggestions

- Smart alerting (no noise)

name: devops-monitor version: 1.0.0 description: AI-powered DevOps monitoring with automated log analysis and alerting

credentials:

• name: GITHUB_TOKEN required: false description: GitHub personal access token
• name: PROMETHEUS_URL required: false description: Prometheus endpoint for metrics
• name: GRAFANA_API_KEY required: false description: Grafana API key
• name: SLACK_WEBHOOK required: false description: Slack webhook for alerts
• name: PAGERDUTY_KEY required: false description: PagerDuty integration key
• name: OPENAI_API_KEY required: true description: OpenAI for AI analysis

inputs:

• name: sources type: array default: [github, logs] description: "Data sources: github, logs, prometheus, uptime"
• name: repos type: array default: [] description: GitHub repos to monitor (owner/repo format)
• name: log_paths type: array default: [] description: Log file paths to analyze
• name: alert_threshold type: string default: high enum: [low, medium, high, critical] description: Minimum severity for alerts
• name: lookback_hours type: integer default: 1 description: Hours of data to analyze

outputs:

• name: status type: object description: "Overall status: {healthy, warnings, errors}"
• name: issues type: array description: "Detected issues: {severity, source, description, suggestion}"
• name: workflows type: array description: GitHub workflow statuses
• name: metrics type: object description: Key metrics summary
• name: alerts_sent type: array description: Alerts that were dispatched
• name: metadata type: object description: Monitoring metadata

backends:

• name: openai type: remote requires: [OPENAI_API_KEY]

components: inputs: [stdin, http_server, http_client] processors: [mapping, http, openai_chat_completion, branch, log] outputs: [http_client, stdout, sync_response]