> For the complete documentation index, see [llms.txt](https://docs.lpp-minduniverse.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.lpp-minduniverse.org/lingua-pactum-protocol-lpp-documentation/11.-public-proof-and-validation/11.7-public-proof-v2.0-controlled-technical-evidence.md).

# 11.7 Public Proof v2.0 — Controlled Technical Evidence

The canonical public evidence repository is:

> **LPP Public Proof**

Current release:

> **Public Proof v2.0 — Controlled Technical Evidence Release**

Release tag:

> `public-proof-v2.0`

Public Proof v2.0 upgrades the repository from primarily architecture / provenance-oriented disclosure into a bounded technical inspection layer for the current Admission Kernel MVP. The published release identifies the final MVP source archive through SHA-256 and exposes schemas, scenario evidence, signed artifacts, a public verification key, a standalone verifier, and a machine-readable validation manifest.

### Source Identity

The release is bound to the final MVP source archive by:

> **SHA-256**\
> `84f901d8d170b5f4110149ce287ab1d722749581ac922151c92a578488bce2d5`

The full private source archive is not part of the public repository.

The hash provides source identity binding without transferring the private runtime itself.

***

### Public Proof v2.0 Includes

The release exposes bounded technical evidence including:

* public JSON Schemas derived from the final MVP,
* current operational reason-code mappings,
* six execution-scenario evidence sets,
* observed scenario outputs,
* signed synthetic MVP evidence,
* signed Admission Artifacts,
* public CP verification material,
* Artifact Verification / Audit Reconstruction material,
* raw validation outputs,
* a machine-readable validation manifest,
* and a standalone public evidence verifier.

The live repository now contains dedicated `docs`, `evidence`, `examples`, `schemas`, `tools`, and `validation` areas rather than functioning only as an architecture/provenance record.

***

### Current Release Results

| Evidence Layer               | Current Release Result       |
| ---------------------------- | ---------------------------- |
| Application test inventory   | 26 tests                     |
| Compatibility execution      | **26/26 PASS**               |
| Standalone core verification | **49/49 PASS**               |
| Core execution scenarios     | **6/6 observed as expected** |
| Public evidence verifier     | **113/113 PASS**             |

These results measure different evidence layers.

They should not be combined into a single aggregate validation score.

***

### Compatibility-Test Environment Boundary

The 26-test application suite was executed in the release-building environment through a minimal Flask-compatible test shim because the release environment did not contain the Flask package.

The compatibility layer implemented only the APIs needed by the application test contract.

Therefore:

> **26/26 PASS**

supports the application logic exercised by those tests.

It does not establish:

* independent validation of Flask,
* production HTTP deployment behavior,
* or production networking behavior.

The standalone 49-check core verifier does not depend on that compatibility shim and exercises core invariants directly against the final source package.

This environment distinction should remain visible.

***

### Public Evidence Recheckability

An external reader can download the disclosed repository and execute the public evidence verifier.

Current published result:

> **113/113 PASS**

This means the disclosed evidence package can be mechanically rechecked.

It does not mean that the private Admission Kernel runtime has been independently reproduced.

Therefore:

> **Evidence re-verification ≠ implementation reproduction**

***

### Public / Private Boundary

#### Public

Current public materials include:

* bounded schemas,
* reason-code semantics,
* execution-scenario definitions,
* captured results,
* synthetic signed evidence,
* public verification material,
* manifests,
* and the public verifier.

#### Not Public

The release does not expose:

* the complete runnable private source package,
* private signing keys,
* production IAM configuration,
* production key-management architecture,
* complete enterprise integration,
* complete internal test harness,
* production deployment topology,
* or proprietary implementation coupling outside the disclosed boundary.

***

### Disclosure-Taxonomy Clarification

The phrase:

> **Controlled Technical Evidence Release**

is the title of the public release model.

It must not be confused with the GitBook disclosure class:

> **P2 — Controlled Technical**

which refers to material that is not generally public.

Public Proof v2.0 itself contains public material.

Where the GitBook disclosure taxonomy is applied, individual released artifacts are classified according to their actual disclosure status rather than by the word “Controlled” in the release title.

***

### Known Implementation Limitation

Source inspection of the final MVP identified a mutable in-memory Authority Record snapshot aliasing risk after later revocation.

A previously created in-memory representation may reference a live Authority Record object that is subsequently mutated.

The persisted append-only JSONL evidence is serialized at creation time and preserves the earlier snapshot.

The current Public Proof reconstruction path therefore treats persisted evidence as the canonical captured evidence source.

The static release evidence does not claim that the mismatch is reproduced in every captured run.

Recommended implementation repair:

> store immutable or deep-copied snapshots before inserting them into the in-memory artifact chain.

This is an implementation-level limitation.

It does not alter LPP protocol semantics.