Example Configs

Example TOML configurations and advanced quickstarts for AURA agents.

New to AURA? Start with the Quickstart — it gets a full stack running in under a minute. Come back here when you’re ready to customize or try orchestration mode.

Reference Configuration

reference.toml is the fully annotated config with every option documented. Use it as a starting point:

cp examples/reference.toml config.toml
# Edit config.toml with your API key and settings
cargo run --bin aura-web-server

If encountering issues and more verbose debugging output is necessary, add the —verbose flag for the running binary

export OPENAI_API_KEY="sk-..."
CONFIG_PATH=examples/minimal/openai.toml cargo run --bin aura-web-server -- --verbose

Minimal Configs

Bare-minimum configs to get running with a single LLM provider. No MCP tools — just add an API key.

export OPENAI_API_KEY="sk-..."
CONFIG_PATH=examples/minimal/openai.toml cargo run --bin aura-web-server

File	Provider	API key needed?
`openai.toml`	OpenAI	Yes
`anthropic.toml`	Anthropic	Yes
`bedrock.toml`	AWS Bedrock	AWS credentials
`gemini.toml`	Google Gemini	Yes
`ollama.toml`	Ollama (local)	No

Complete Agent Configs

Full agent compositions that combine an LLM provider with real MCP tool servers and tailored system prompts. Each file documents its prerequisites and required environment variables.

File	Description
`devops-assistant.toml`	GitHub MCP
`incident-response-mezmo.toml`	PagerDuty + Mezmo
`incident-response-datadog.toml`	PagerDuty + Datadog
`kubernetes-sre.toml`	K8s MCP + Prometheus

Serving Multiple Agents

Point CONFIG_PATH at a directory to serve every .toml file as a selectable agent:

CONFIG_PATH=examples/complete/ cargo run --bin aura-web-server

Clients discover agents via GET /v1/models and select one with the model field in chat requests. Each agent is identified by its alias (if set) or name.

Advanced Quickstarts

Self-contained Docker Compose setups for specific use cases. Each has its own guide with step-by-step instructions.

Quickstart	Description
Orchestration — Math MCP	Multi-agent orchestration with coordinator/worker architecture and a math tool server
Kubernetes SRE	AI-powered SRE agent on KIND with Kubernetes and Prometheus MCP servers

Getting Started

Configuration

Features

CLI Reference

Web Server Reference

Reference Configuration

Minimal Configs

Complete Agent Configs

Serving Multiple Agents

Advanced Quickstarts

​Reference Configuration

​Minimal Configs

​Complete Agent Configs

​Serving Multiple Agents

​Advanced Quickstarts

Reference Configuration

Minimal Configs

Complete Agent Configs

Serving Multiple Agents

Advanced Quickstarts