Verification

Check the record. Understand the result.

Each verification check answers a specific question. Orvessian keeps those answers separate so you can see what the evidence supports.

What can you check?

Record integrity. The disclosed record and opening reproduce the recorded commitment.

Log inclusion. The commitment belongs to the history represented by a particular checkpoint.

Checkpoint signature. The checkpoint was signed by a key accepted under the verifier’s trust policy.

History consistency. A later checkpoint extends the earlier history being compared.

Independent timestamp, when present. The checkpoint existed by an attested time, subject to the authority’s trust and accuracy assumptions.

Verify outside the workspace

Download the evidence for a selected record. Use separately trusted keys and policies to check the available proof material without relying on the dashboard’s verdict.

If the original payload is no longer available, the export identifies which checks can still be performed. Missing evidence is reported explicitly.

How the evidence works

Orvessian accepts the structured record into its database and creates a cryptographic commitment. That commitment enters an ordered log.

A signed checkpoint identifies a particular state of the log. An inclusion proof connects the record’s commitment to that checkpoint. Comparing checkpoints can establish that one history extends another.

Checkpoints retained independently provide an additional reference for detecting changes to previously observed history.

Understand the boundaries

A valid signature identifies an accepted signing key. It does not automatically authenticate the original agent or model.

An inclusion proof does not show that every relevant event was submitted. A timestamp does not establish the exact time of the underlying action.

Read the technical paper →