ScopeAudit links the agreed scope to delivered code, tests, documentation, changes, and supplier claims—then produces evidence-backed findings, remediation, and a qualified acceptance recommendation.
We check whether an agency actually delivered the software its client paid for.
How the products work together
Why ScopeAudit
Commissioned work crosses organisational boundaries. The supplier holds most delivery evidence, while the client carries the acceptance decision. A Done ticket and green CI do not, by themselves, prove conformity with the agreed scope.
Turn the accepted scope into traceable requirements.
Preserve what the supplier says is complete.
Connect code, tests, documents, changes, and explicit absence.
Derive findings from a validated evidence graph.
State blockers and the evidence required for remediation.
The product system
ScopeGraph establishes what the delivery evidence says. ScopeAudit applies acceptance logic to that evidence and explains what should happen next.
Evidence layer
A typed evidence map of the commissioned delivery—not a decorative diagram.
Requirements, supplier assertions, code, tests, documents, changes, and explicit absences become inspectable nodes. Source-backed relationships preserve what is connected, why it is connected, and how confident the audit can be.
Decision layer
A milestone acceptance audit built from the evidence graph.
Deterministic rules test delivery claims against the agreed baseline. ScopeAudit consolidates the resulting issues into material findings, required remediation, and a qualified recommendation for the client’s acceptance decision.
Product boundary
Product v0 is a working vertical-slice prototype using a synthetic commissioned-software delivery package. It demonstrates the audit workflow, evidence graph, deterministic rule model, and acceptance report. Production integrations, arbitrary repository analysis, and human-reviewed live audits are subsequent work.