RESOURCES · DOCS · AGENT STUDIO · OPEN SOURCE

Debug, replay, and ship MCP and A2A agents.

Agent Studio Open runs on your own machine. The traces, the keys, and the exports stay local. No pooled data, nothing to send us. Read the docs, then run it yourself.

Open agentstudio.auraone.ai Start in 60 seconds

DOCS SURFACE

Agent Studio Open

LOCAL-FIRST

ConnectComposeTracesReplayObserveShip

COMPOSE · TOOL CALL

{
  "tool": "refund_order",
  "order_id": "ORD-1042",
  "dry_run": true
}

TOOLS

TRACE

.ast

P95

812ms

TRACE READOUT

YOU RUN IT

Local-first

No account required. Open source you run on your own infrastructure. Trace data stays on the machine until you choose to export it.

PROTOCOLS

MCP · A2A · OTEL

stdio, SSE, HTTP, WebSocket. Agent cards. OpenTelemetry GenAI spans.

EXPORTS

CI-ready

GitHub Actions, JUnit, PR comments, trace cards, intake packets.

INSTRUMENT READOUT · LOCAL TRACE STORE

DOCUMENTATION MAP

Find the page. Ship the workflow.

Pick the surface you are debugging. Each guide names the connection mode, what to inspect, what to record, and what artifact to export. Versioned with the platform.

START HERE

4 MIN

60-second quickstart

Connect a sample MCP server, record a tool call, replay it, and export a regression artifact.

Read guide →

MENTAL MODEL

7 MIN

Concepts

Connections, sessions, replays, regression banks, trace cards, and intake packets.

Read guide →

COOKBOOK

12 MIN

MCP cookbook

Worked MCP debugging recipes for stdio, SSE, HTTP, WebSocket, risk linting, and replay exports.

Read guide →

COOKBOOK

8 MIN

A2A cookbook

Five worked examples for agent cards, capability negotiation, task lifecycle, and failure paths.

Read guide →

OBSERVE

9 MIN

OTEL ingest cookbook

Import OpenTelemetry GenAI spans, Phoenix JSON, OTLP HTTP JSON/proto, and OTLP gRPC into local traces.

Read guide →

SETTINGS

6 MIN

Provider keys

Configure OpenAI, Anthropic, Google, and Ollama without sending keys to AuraOne.

Read guide →

REFERENCE

10 MIN

CLI reference

The agentstudio binary for connect, record, replay, compare, risk-scan, A2A, and export.

Read guide →

EDITOR

6 MIN

VS Code extension

Discover MCP servers from a workspace, inspect manifests, compose calls, and deep-link to desktop.

Read guide →

WEB

6 MIN

Browser edition

Use Agent Studio Open from agentstudio.auraone.ai with remote transports and local browser storage.

Read guide →

DEEP DIVE

13 MIN

Architecture

Tauri shell, Rust core, Python sidecar, trace store, protocol adapters, and export pipeline.

Read guide →

FAQ

7 MIN

FAQ

Common product, protocol, privacy, commercial-roadmap, and launch questions.

Read guide →

TRUST

8 MIN

Privacy and telemetry

What stays local, what telemetry can be enabled, what exports contain, and how redaction works.

Read guide →

SECURITY

8 MIN

Sandbox guide

Run untrusted stdio MCP servers with declared paths and limited network/file access.

Read guide →

HELP

10 MIN

Troubleshooting

Connection failures, malformed manifests, Python sidecar issues, OTLP errors, and replay nondeterminism.

Read guide →

FIRST COMMANDS

The shortest path to a replay.

Install, connect to a sample server, record a session, replay it across providers, and export the artifact.

agentstudio connect --stdio "python server.py"
agentstudio record --session refund-path
agentstudio replay refund-path.ast --compare anthropic,openai
agentstudio export refund-path.ast --github-action

AGENT STUDIO OPEN

Run it yourself. Own the trace.

Local-first by design. The trace, the keys, the replay, and the export stay on the machine until the team decides otherwise. You run the tool. You keep what you build.

Open agentstudio.auraone.ai Start in 60 seconds