Skip to content

Run a cloud audit

A cloud audit drives a real screen reader against your saved journeys on a dedicated virtual machine. It is asynchronous: you start it, poll it, and read the report when it finishes.

Use a project id to find or create saved journeys. Starting a saved journey takes its journey id.

list_journeys(project_id)

If the project has none, create one — a journey is a named user flow, described in ordered steps:

create_journey(
project_id,
name: "Checkout",
description: "Buy one item as a guest",
steps: ["Open the product page", "Add to cart", "Complete checkout as a guest"]
)
start_agentic_audit(journeyIds: ["jrn_…"])

For a project-wide audit, name projectId and omit journeyIds and targetId. The call returns the run id of each journey, queued from the start:

Cloud audit started. Run ID: run_… (VM: …)
The run is queued while a dedicated VM boots — this takes ~4–5 minutes before it starts running.
get_audit_progress(organization_id, run_id)

While the VM boots, the run is queued. That is the normal state, not an error — keep polling. If the VM never starts, the run ends failed with the reason never_started.

The response is the run itself, with a status and violation counts. Its id is the run id — the one every other tool takes too.

Keep polling while the status is queued or running. The run is finished on any of completed, completed_with_errors, failed, or interrupted — note that completed_with_errors means partial coverage with a readable report, so do not wait for completed alone.

Once the run has finished (completed or completed_with_errors):

get_agentic_audit_report(runId: "<run id>")

The report’s tickets are the issues the run found, one per distinct issue, each with the element it sits on, the criterion it maps to, its impact, its status, and — where one was captured — a screenshotKey. An issue found again, in this run or an earlier one, stays one ticket.

get_violation_screenshot(organizationId: "<organization ID>", key: "<the ticket's screenshotKey>")

The field in the report is screenshotKey; the argument here is key. Pass the one into the other.

This returns an image, not text, so an agent that supports images can look at the failure directly.

cancel_audit(organization_id, run_id)

The stop is a request, not an immediate kill. Poll get_audit_progress until the status reads interrupted to confirm it actually stopped.

list_project_run_history(organization_id, project_id, first, after)

This merges two kinds of entry: the cloud audits Binclusive drove (AgenticRun) and the static scans your own CI reported through b8e ci (CliScan). Each entry names its type in __typename. The list is paginated because CI history grows a row per push — see Page through large results.

For the exact arguments of every tool here, see the audit tool reference.