Aura

v1.0.0Coding Agentsstable

A production-ready framework for composing AI agents from declarative TOML configuration, with MCP tool integration, RAG pipelines, and an OpenAI-compatible web API.

ai-agent-frameworkai-agentsai-sreaiopsdevops
Share:
68
Stars
0
Downloads
0
Weekly
0/5

What is Aura?

Aura is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to production-ready framework for composing ai agents from declarative toml configuration, with mcp tool integration, rag pipelines, and an openai-compatible web api.

A production-ready framework for composing AI agents from declarative TOML configuration, with MCP tool integration, RAG pipelines, and an OpenAI-compatible web API.

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • A production-ready framework for composing AI agents from de

Use Cases

Declarative AI agent framework with TOML config
RAG pipelines and OpenAI-compatible API
mezmo

Maintainer

LicenseApache-2.0
Languagerust
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx aura

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Aura

Aura is a production-ready, Rust-based framework for composing AI agents from declarative TOML configuration files. It exposes an OpenAI-compatible REST API, supports multiple LLM providers (OpenAI, Anthropic, Bedrock, Gemini, Ollama, OpenRouter), and integrates MCP tools over HTTP, SSE, and STDIO transports. Aura includes multi-agent orchestration with coordinator/worker DAGs, vector search via Qdrant, scratchpad context management for large tool outputs, and OpenTelemetry tracing — making it suitable for enterprise DevOps, SRE, and AIOps workflows where governed, observable agent pipelines are required.

Prerequisites

  • Rust toolchain (for building from source) OR Docker and Docker Compose (for the containerized quick start)
  • An LLM provider API key — e.g., `OPENAI_API_KEY` for OpenAI or compatible providers
  • An MCP-compatible client or any OpenAI-compatible client to consume the Aura API
  • Optional: Qdrant running locally for vector search / RAG pipeline support
1

Quick start with Docker Compose

The fastest way to run Aura is with Docker Compose. Copy the example environment file, set your API key, and start the stack.

cp .env.example .env
# Edit .env and set OPENAI_API_KEY=your-key
docker compose up -d
2

Build from source with Rust (alternative)

If you prefer to build locally, install Rust, clone the repo, set your API key, and run the web server binary.

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
export OPENAI_API_KEY="your-api-key"
cargo run --bin aura-web-server
3

Create a minimal agent TOML configuration

Define your agent in a TOML file. The `alias` is used as the `model` field in OpenAI-compatible API calls to select this agent.

# config.toml
[agent]
name = "Assistant"
alias = "my-assistant"
system_prompt = "You are a helpful assistant."
turn_depth = 2

[agent.llm]
provider = "openai"
api_key = "{{ env.OPENAI_API_KEY }}"
model = "gpt-4o"
context_window = 128000
4

Add an MCP server to your agent

Extend the TOML config to connect an MCP server. Aura supports HTTP streamable, SSE, and STDIO MCP transports.

[mcp.servers.my_tools]
transport = "http_streamable"
url = "http://localhost:8081/mcp"
5

Start the server and verify it is healthy

With your config in place, start Aura and hit the health endpoint to confirm it is running.

CONFIG_PATH=config.toml cargo run --bin aura-web-server
# In another terminal:
curl http://localhost:8080/health
6

Send a chat completion request

Use the OpenAI-compatible endpoint to chat with your agent. Set the `model` field to the agent's `alias` from your TOML config.

curl -X POST http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model": "my-assistant", "stream": true, "messages": [{"role": "user", "content": "Hello"}]}'

Aura Examples

Client configuration

Because Aura exposes an OpenAI-compatible API, configure your MCP client to connect to the Aura web server. Set the `OPENAI_API_KEY` env var and point the base URL at Aura.

{
  "mcpServers": {
    "aura": {
      "command": "cargo",
      "args": ["run", "--bin", "aura-web-server"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key",
        "CONFIG_PATH": "/absolute/path/to/config.toml",
        "HOST": "127.0.0.1",
        "PORT": "8080"
      }
    }
  }
}

Prompts to try

These prompts work via the OpenAI-compatible API or through any MCP client connected to your Aura instance.

- "List all available agents in this Aura instance"
- "Using the devops-agent, describe the current state of my Kubernetes cluster"
- "Run a multi-agent research task: have the researcher gather data and the summarizer produce a report"
- "Query the vector store for documents related to incident response playbooks"
- "Show me the OpenTelemetry traces for the last agent run"

Troubleshooting Aura

Server fails to start with a config parsing error

Validate your TOML file syntax at https://www.toml-lint.com/ or with `taplo lint config.toml`. Ensure the `alias` field is unique across all agents and that environment variable references use the `{{ env.VAR_NAME }}` syntax.

MCP tools are not discovered or calls fail

Verify the MCP server URL is reachable from the host running Aura: `curl http://localhost:8081/mcp`. Ensure the `transport` value in the TOML matches what the MCP server supports (http_streamable, sse, or stdio).

Docker Compose starts but the agent returns LLM errors

Check that `OPENAI_API_KEY` (or the relevant provider key) is set in your `.env` file and that the key is valid. Run `docker compose logs aura` to see the full error message from the LLM provider.

Frequently Asked Questions about Aura

What is Aura?

Aura is a Model Context Protocol (MCP) server that production-ready framework for composing ai agents from declarative toml configuration, with mcp tool integration, rag pipelines, and an openai-compatible web api. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Aura?

Follow the installation instructions on the Aura GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Aura?

Aura works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Aura free to use?

Yes, Aura is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "aura": { "command": "npx", "args": ["-y", "aura"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Aura?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides