STEP 01
WHAT WE OPEN
Connect
Start with a local stdio server, a remote SSE or HTTP MCP endpoint, an A2A card, or an imported OTEL trace.
→OPEN · AGENT STUDIO · MCP AND A2A, LOCAL
The debug loop, on your laptop.
A local-first IDE for MCP and A2A agents. Connect, inspect, record, replay, compare model behavior, ingest OTEL spans, and export regression suites to CI.
SESSION · refund_order
TOOLS
18
REPLAY DIFF
3
P95
812ms
CONNECT
MCP and A2A
stdio, SSE, HTTP, WebSocket. Cards, manifests, tools, risk findings.
REPLAY
Deterministic
Record the path. Mock the tool outputs. Compare across models.
SHIP
CI artifacts
GitHub Action, JUnit, PR comment, trace card, intake packet.
HOW IT WORKS
Three steps. No orchestration framework.
Connect the server. Replay the path. Ship the regression. The trace stays on disk.
STEP 02
WHAT WE LOCK
Replay
Record the agent path. Replay it with mocked tool outputs. Compare per-turn and per-tool across model versions.
→STEP 03
WHAT WE EXPORT
Ship
Generate a GitHub Action, JUnit report, PR comment, trace card, or AuraOne intake packet for the next reviewer.
WHAT COMES OUT
Every run leaves a portable artifact.
Repo-ready files. No hosted account. The next reviewer can pick the run back up from disk.
01
Trace cards
Portable Markdown and JSON for one agent run: tools, retries, data touched, outcome, failure mode.
↳ ARTIFACT
02
Regression suites
Every failed tool call becomes a deterministic replay the next release must clear.
↳ ARTIFACT
03
GitHub Actions
A drop-in workflow file that runs the replay set on every push and posts findings to the PR.
↳ ARTIFACT
04
JUnit reports
Standard XML for the CI dashboard your team already runs. No new viewer required.
↳ ARTIFACT
05
Intake packets
Signed .auraonepkg with a privacy preview before handoff to AuraOne reviewers.
↳ ARTIFACT
RELATED OPEN SURFACES
Next to this in AuraOne Open.
RUBRIC STUDIO OPEN
The IDE for the rubric.
Local, file-based, git-friendly criterion-level authoring.
See the page →ROBOTICS STUDIO OPEN
Review teleop and VLA datasets, on disk.
Scrub synchronized sensor streams. Cluster failures. Export reviewed subsets.
See the page →OPEN V2
Trust gates for agentic and embodied AI.
Twelve installable packages. mcp-risk-linter, a2a-contract-test, tool-call-replay, more.
See the page →AGENT STUDIO OPEN
Open the agent. Own the trace.
The debug loop, your tool calls, your replay artifacts — all on disk. No telemetry by default.