feat(blueprints): add autonomous incident triage with multi-provider LLM fallback and Slack escalation - #342

What

Adds an enterprise-grade production workflow blueprint: Autonomous Incident Triage with Multi-Provider LLM Fallback and Slack Escalation (flows/autonomous-incident-triage-with-fallback.yaml).

Blueprint

• File: flows/autonomous-incident-triage-with-fallback.yaml
• ID: autonomous-incident-triage-with-fallback
• Namespace: company.team
• Tags: AI, Infrastructure, Business

What it does

1. Inbound Incident Event Handling: Listens via Webhook and scheduled cron triggers for alert payloads containing error stack traces, service names, and error counts.
2. Three-Tier LLM Failover Architecture:
   • Tier 1 (Primary): OpenAI gpt-4o-mini for high-accuracy root-cause analysis and severity scoring.
   • Tier 2 (Fallback): Google Gemini gemini-1.5-flash with zero-latency failover when OpenAI rate-limits or errors.
   • Tier 3 (Emergency Heuristic Engine): Rule-based Python regex AST analyzer guaranteeing zero dropped incidents even under complete cloud provider outages.
3. Automated Blast Radius Evaluation: Computes impacted downstream services and SLO degradation.
4. Human-in-the-Loop Approval Gate: Uses Kestra Pause task (PT1H timeout) for critical/P0 severity incidents requiring human sign-off before automated mitigation runs.
5. Slack Ops Escalation & Audit Trail: Posts structured triage cards to #production-incidents and logs end-to-end execution metadata.

Prerequisites

• OpenAI API Key (OPENAI_API_KEY) and/or Google Gemini API Key (GEMINI_API_KEY).
• Slack Incoming Webhook URL (SLACK_WEBHOOK_URL).

Validation

• PyYAML syntax validated.
• Blueprint tag workflow (.github/workflows/validate-blueprint-tags.yml) verified: valid tags, metaDescription length <= 160 characters.
• 100% emoji-free output.