aura
AURA is a production-tested SRE agent platform you can deploy in minutes. AURA handles the guardrails, APIs, state management, streaming, and failure handling
What is aura?
AURA is a production-tested SRE agent platform you can deploy in minutes. AURA handles the guardrails, APIs, state management, streaming, and failure handling required to put AI to work safely on production infrastructure.
Overview of aura
What it is
AURA is a production-tested SRE agent platform that can be deployed in minutes. It handles guardrails, APIs, state management, streaming, and failure handling required to connect AI models to production tools within operator-defined boundaries.
Key features
- Operate inside your own security boundary, including air-gapped and highly regulated environments.
- Define the entire agent system in reviewable TOML configuration that can be versioned and reviewed.
- Use different model providers (OpenAI, Anthropic, Bedrock, Gemini, Ollama, OpenRouter) without rebuilding workflows.
- Put sensitive actions behind human approval with webhook or in-conversation approval, fail-closed on denial or timeout.
- Trace every model, tool, and orchestration decision via OpenTelemetry exports.
Use cases
AURA is used for investigating incidents, such as a payment failure, by correlating Mezmo traces and logs, Prometheus latency, and Kubernetes deployment history. It is also used for production operations by SRE teams, both inside Mezmo and externally. The platform powers thousands of agent sessions each month in Mezmo’s hosted observability platform.
Getting started
Install AURA on Linux or macOS with the install script: bash curl -fsSL https://raw.githubusercontent.com/mezmo/aura/main/scripts/install.sh | bash Create a ready-to-run local agent: bash aura init Start the agent: bash aura
License and pricing
AURA is licensed under the Apache License, Version 2.0. It is open source.
Based on the project's public README and repository metadata.