Quick Start

OpenSRE runs as a set of Docker Compose services. You can have a fully functional AI SRE platform running locally in under 5 minutes.

Prerequisites

  • Docker and Docker Compose installed
  • An Anthropic API key — OpenSRE's agent runtime is built on the Claude Agent SDK and calls Anthropic directly by default. You only need ANTHROPIC_API_KEY.
  • Git to clone the repository

Installation

1. Clone the repository

git clone https://github.com/swapnildahiphale/OpenSRE.git
cd OpenSRE

2. Configure environment

Copy the example env file and set your key:

cp .env.example .env

# .env
ANTHROPIC_API_KEY=sk-ant-your-api-key

That's the only required variable to get started. See Configuration for the full list of optional settings.

Want to route through OpenRouter or another provider instead? OpenSRE ships an optional LiteLLM proxy for that — see Use a different LLM provider below. It's off by default; direct Anthropic is the supported path for make dev.

3. Start all services

make dev

This starts the core stack:

ServicePortDescription
PostgreSQL5433Config data + investigation run traces
config-service8081Configuration API
Neo4j7475 (HTTP), 7688 (Bolt)Episodic memory + knowledge graph
sre-agent8001Investigation agent (Claude Agent SDK)
web-ui3002Admin console

4. Generate a token and sign in

make dev boots config-service in local mode, which auto-seeds one organization (local) and one team (default) from config_service/config/local.yaml — there's no separate "create an org" step for local development. Mint a team token for that org/team:

make e2e-token

This calls config-service's admin API (using the default admin token, local-admin-token) and prints a team token. Now open http://localhost:3002 — the console opens on a sign-in screen. Paste the team token there to log in.

Want admin access instead? Paste local-admin-token (the default ADMIN_TOKEN) at the same sign-in screen. That unlocks Admin — org tree, token management, and SSO config — where you can mint and revoke tokens visually instead of re-running make e2e-token. See Configuration for token scopes and the SSO setup.

Trigger Your First Investigation

In the web console, you can trigger a test investigation:

  1. Click New Investigation in the top bar (or press ⌘I)
  2. Enter an alert description, for example: High error rate on payments-service: 5xx errors spiked to 15% in the last 10 minutes
  3. Click Investigate
  4. Watch as OpenSRE's agents gather context, investigate in parallel, and produce a report

The first investigation may take a few minutes depending on which skills are invoked and how deep the agent decides to go.

Add Slack Integration (Optional)

To receive alerts and investigations via Slack:

# Add to your .env
SLACK_BOT_TOKEN=xoxb-your-bot-token
SLACK_APP_TOKEN=xapp-your-app-token

# Start with Slack bot
make dev-slack

Use a Different LLM Provider (Optional)

By default make dev calls Anthropic directly. If you'd rather route through OpenRouter or another provider, start the optional LiteLLM proxy alongside the core stack and point the agent at it:

# Add to your .env
OPENROUTER_API_KEY=your-openrouter-api-key-here
ANTHROPIC_BASE_URL=http://litellm:4000

# Start the core stack plus the LiteLLM profile
docker compose --profile litellm up

See Configuration for the full litellm_config.yaml setup.

Next Steps