Connect GitHub, Datadog, Kubernetes, PagerDuty and other DevOps tools into one queryable system.
Ask a single question during an incident and let agents trace the root cause across every connected tool.
Run a local demo tenant to show stakeholders how agentic incident response could work.
Deploy the platform to production with Helm charts and Terraform modules for AWS, GCP or Azure.
| anway-dev/anway | adguardteam/ruleseditor | amazing-things/open-controller | |
|---|---|---|---|
| Stars | 7 | 7 | 7 |
| Language | TypeScript | TypeScript | TypeScript |
| Last pushed | — | 2026-07-01 | — |
| Maintenance | — | Active | — |
| Setup difficulty | hard | easy | moderate |
| Complexity | 4/5 | 2/5 | 3/5 |
| Audience | ops devops | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires Docker Compose for Postgres, Redis and Neo4j, plus an LLM API key for chat.
Anway is a beta platform that positions itself as a central hub connecting all the tools a software team already uses, things like GitHub, Datadog, Linear, Kubernetes, Loki, Prometheus, Jira, ArgoCD, PagerDuty, Terraform and various cloud providers. The idea is that instead of an engineer, SRE, product manager or business analyst jumping between a dozen dashboards, everyone can ask a single question and get an answer that pulls context from across the whole toolchain, then take action through one governed system. The typical flow described is that a signal, like an alert or incident, fires, someone asks a question once, and background agents trace the root cause by looking across every connected tool. From there the system can carry out a gated action, meaning a human still approves anything sensitive, and everything is recorded in an audit trail. The project is split across two repositories, this one holds the actual platform, including the web interface, gateway, agent logic, command line tool and connectors, while a separate repo holds the cloud testing environment built with Terraform and Kubernetes manifests. Technically the system is a set of connected services. A Next.js web app talks to a Fastify gateway that handles authentication, permissions and auditing, which in turn talks to a Python FastAPI service that runs the AI reasoning and keeps a memory graph of past events. Data is stored in PostgreSQL with vector and graph extensions, Redis handles session memory and background job queues, and Neo4j stores the episodic event graph. To run it locally you need Node.js, pnpm, Docker and Python, then you start the supporting infrastructure with Docker Compose, run database migrations and a seed script that fills in sample services and incidents, and finally start the gateway and web app in separate terminals. There is also a demo mode that spins up a single pre populated tenant so anyone can log in without credentials, meant only for live demos, never for production use. The project is released under the MIT license and is still actively developed, so its APIs and data structures may still change.
A beta platform that connects GitHub, Datadog, Kubernetes and other DevOps tools into one AI-queryable surface for incident response.
Mainly TypeScript. The stack also includes TypeScript, Next.js, Fastify.
Free to use, modify and redistribute for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly ops devops.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.