$ -weight: 500;">git clone https://github.com/Arvo-AI/aurora.-weight: 500;">git cd aurora make init make prod-prebuilt -weight: 500;">git clone https://github.com/Arvo-AI/aurora.-weight: 500;">git cd aurora make init make prod-prebuilt -weight: 500;">git clone https://github.com/Arvo-AI/aurora.-weight: 500;">git cd aurora make init make prod-prebuilt - Multi-agent architecture with parallel hypothesis testing - Formulates multiple theories per incident, deploys sub-agents to investigate each simultaneously - Correlates alerts across services and dependencies - Constructs causal timelines linking code changes, infra events, and telemetry - Generates root cause analysis with confidence scores - Human-in-the-loop approval gates before automated actions - Per-customer fine-tuned models - Multi-agent architecture via LangGraph with dynamic tool selection (30+ tools) - Correlates alerts across services and dependencies (AlertCorrelator + Memgraph graph) - Constructs investigation timelines linking deployments, infra events, and telemetry - Generates structured RCA with evidence citations and remediation steps - Human-in-the-loop for write/destructive actions — read-only commands run automatically - Executes -weight: 500;">kubectl, aws, az, gcloud commands in sandboxed Kubernetes pods (non-root, read-only filesystem, capabilities dropped, seccomp enforced) - Queries cloud APIs directly — AWS (STS AssumeRole), Azure (Service Principal), GCP (OAuth), OVH, Scaleway - Traverses Memgraph infrastructure dependency graph for blast radius - Searches Weaviate knowledge base (vector search over runbooks and past incidents) - Works with any LLM provider — choose your own model - AWS and GCP confirmed - Azure is not listed on their integrations page - Kubernetes support confirmed - Deploys an on-premise "satellite" agent as a secure gateway — core platform runs in Resolve's cloud - AWS, Azure, GCP, OVH, Scaleway — all five with native authentication - Deep Kubernetes integration via outbound WebSocket -weight: 500;">kubectl-agent - Fully self-hosted — Docker Compose or Helm chart - No data leaves your environment - Monitoring: Grafana, Datadog, Splunk, Prometheus, Dynatrace, Elastic, Chronosphere, Kloudfuse, OpenSearch - Infrastructure: Kubernetes, AWS, GCP - Code: GitHub - Chat: Slack - Knowledge: Notion - Custom: MCP, APIs, Webhooks - Total: ~17+ confirmed - Monitoring: PagerDuty, Datadog, Grafana, New Relic, Netdata, Dynatrace, Coroot, ThousandEyes, BigPanda, Splunk - Cloud: AWS, Azure, GCP, OVH, Scaleway - Infrastructure: Kubernetes, Terraform, Docker - CI/CD: GitHub, Bitbucket, Jenkins, CloudBees, Spinnaker - Docs: Confluence, Jira, SharePoint - Network: Cloudflare, Tailscale - Communication: Slack - Total: 25+ confirmed - Learns from runbooks, wikis, chats, and historical incidents - Builds a knowledge graph of infrastructure components - Captures tribal knowledge from production systems - Per-customer fine-tuned models that improve from feedback (thumbs up/down) - Built-in Weaviate vector store for semantic search over runbooks, postmortems, and documentation - Memgraph infrastructure dependency graph maps relationships across all cloud providers - Learns from past investigations stored in the knowledge base - Automatic JIRA ticket updates during investigation - Enterprise support with SLAs - Available on AWS Marketplace - Azure, OVH, and Scaleway cloud support - Open source (Apache 2.0) — full codebase auditable - Self-hosted deployment (Docker Compose, Helm) - LLM provider flexibility (OpenAI, Anthropic, Google, Ollama for air-gapped) - Slack incident channel creation and management - PagerDuty, New Relic, BigPanda, ThousandEyes, Coroot integrations - Terraform/IaC state analysis - Bitbucket, Jenkins, CloudBees, Spinnaker integrations - Confluence and SharePoint integration - Network integrations (Cloudflare, Tailscale) - Free — no licensing costs whatsoever - Autonomous AI incident investigation - Multi-agent architecture - Root cause analysis with evidence - AI-suggested code fixes (human-approved PRs) - Infrastructure dependency/knowledge graph - Knowledge base search (runbooks, wikis, past incidents) - Kubernetes investigation - AWS and GCP support - Datadog, Grafana, Splunk, Dynatrace integrations - Slack integration - RBAC and security controls - AI that learns from user feedback - Causal timeline construction with dependency chain mapping - Human-in-the-loop for destructive actions - Per-customer tuning (Resolve.ai via fine-tuned models; Aurora via open source customization) - SOC 2 Type II compliance (Resolve.ai: certified; Aurora: in progress) - No public pricing page - Custom enterprise pricing (contact sales) - No free tier or self--weight: 500;">service signup - Target: large enterprise SRE teams - Free — Apache 2.0, self-hosted - Costs: infrastructure (VM or K8s cluster) + LLM API usage - $0 LLM cost with Ollama local models - No contracts, no sales calls, no per-user pricing - Read every line of code the AI uses to investigate your infrastructure - Self-host with zero data leaving your environment - Use any LLM provider — or run local models for fully air-gapped operation - Modify investigation workflows, add custom tools, fork for your needs - Contribute back to the project - You're a large enterprise company with budget for enterprise AI tooling - Managed fine-tuned models — you want the vendor to handle per-customer model training rather than customizing open source yourself - You need certified compliance today — SOC 2 Type II, HIPAA, GDPR already certified (Aurora's SOC 2 is in progress) - Managed -weight: 500;">service preferred — you don't want to maintain AI infrastructure - Budget matters — you can't justify custom enterprise pricing for AI investigation - Open source is required — you need full transparency into how AI investigates your production systems - Self-hosted is required — compliance, data sovereignty, or air-gapped environments - Multi-cloud breadth — you need Azure, OVH, or Scaleway alongside AWS and GCP - LLM flexibility — you want to choose your own provider or run models locally - You're a startup or mid-market — Resolve.ai has no mid-market pricing - You want a custom integration — the Arvo AI team actively builds custom integrations for companies at no cost. If there's a feature gap, reach out and they'll build it with you. - Aurora vs Traditional Incident Management Tools — Comparison with Rootly, FireHydrant, incident.io - PagerDuty Alternative for Root Cause Analysis — PagerDuty vs Aurora deep dive - Rootly Alternative: Open Source AI Incident Management — Rootly vs Aurora - Aurora Documentation — Full setup and configuration guides - Joined Mar 26, 2026