Credential gated transaction target — 2027-07-20 target

D-19 story. As a relying party, I want the released demo to authorize a Cardano action only while the presented synthetic credential chain is valid under the recorded issuer and revocation state, so that a recorded upstream revocation visibly ends access.

Planned release tag: Cardano KERI v2.0.0 (credential-gate M2). This names the milestone target; intermediate plays use release candidates, and the tag waits for the D-19 release gate and its preprod evidence.

This is a dated review target from the live project card. Not yet playable as the connected target. No confirmed ledger outcome or release acceptance is claimed by this page.

Planned play and evidence

Given Cardano preprod, the documented release artifact and synthetic four-link credential fixtures; when an independent operator uses keripy and ckeri to run the connected setup, submits the valid action, records a QVI revocation and submits the same action again; then the first ledger action succeeds and is readable back, while the second is refused by the on-chain gate; every result is tied to the exact model, scripts, transaction IDs and evidence package.

sequenceDiagram
    participant A as Story actor
    participant M as Accepted model
    participant C as Cardano boundary
    A->>M: Prepare the stated evidence
    M-->>A: Check expected transition and refusal
    A->>C: Submit the connected action when implemented
    C-->>A: Read back the actual result or refusal

The intended positive observation is: An independent operator uses a documented release and Cardano preprod to run the connected valid action, recorded QVI revocation and retry. The relevant refusal is: The on-chain gate refuses the retry after the revocation. These are acceptance criteria, not observed results.

Presenter path: 10–15 minutes when runnable

Time	Presenter action	Evidence to inspect
0–2 min	State the actor's goal and identify all synthetic inputs.	Pin the exact accepted model and release revisions.
2–5 min	Establish the card's given state through its connected prior operations.	Inspect the actual starting outputs and required evidence.
5–8 min	Run the planned positive action.	Compare fresh readback with the bound model transition.
8–11 min	Run the planned negative control.	Attribute the refusal to the actual boundary.
11–15 min	Review receipts and gaps.	Identify transactions, scripts, witnesses, values and uncovered behavior.

Missing interface or receipt

One runnable released script and independent replay, archive/model/fixture hashes, positive and revocation transaction IDs, refusal trace and execution units. This is the narrow target; the full #45 six-step cache, detached-submission, expiry and scoped-override demo remains separate.

The model base visible in this checkout is Cardano KERI commit 0e638fadc987f0cd98f839edd3e3ecd5a09a1b91. It is a source identity for planning, not a claim that the later card is already modeled or accepted. Each claim must bind the accepted model revision for that behavior before promotion. The Lean correspondence register holds affected conflicts. A simulator or source fact cannot replace a confirmed transaction, and no cast of an accepted connected result is available here.

Tracking: First vertical of KERI M2 demo #45, with verifier #31, builder #32, and mirror #392.