Status: controlled usability trial / package path / non-mutating

Controller: GitHub issue #1210.

This exercise tests whether one technically capable operator who did not build Governed Repo can install and use the package from public instructions alone.

It is not a certification, security assessment, production approval, or independent scientific validation.

The operator should:

• not be the implementation author;
• not rely on private NDR ecosystem context;
• not receive step-by-step coaching from the implementation author;
• record any ambiguity, failed command, hidden assumption, or needed clarification instead of silently working around it.

If a blocking defect appears, record it. A defect is useful usability evidence.

Use this exact source commit:

c90c74c04583c5a3f23f6a261cf38cd6922c781c

This merged commit is the canonical protected-main source for the trial. It has the same Git tree as fully tested PR head 77c818ecc5d57e5dd354bc86396dc9e7c5c6be9e, which completed the Governed Repo package workflow successfully on Python 3.10, 3.11, 3.12, 3.13, and 3.14. The merged commit subsequently completed the observed post-merge mainline suite successfully.

The package is not published. Install it from this exact merged source checkout.

You need:

• Git;
• Python 3.10 or newer;
• pip.

Record your operating system and Python version before continuing:

python --version
git --version

git clone https://github.com/ndrorchestration/DGAF-Framework.git
cd DGAF-Framework
git checkout c90c74c04583c5a3f23f6a261cf38cd6922c781c
git rev-parse HEAD

Expected final SHA:

c90c74c04583c5a3f23f6a261cf38cd6922c781c

If the SHA differs, stop and record the mismatch.

python -m venv .venv-governed-repo
.\.venv-governed-repo\Scripts\python.exe -m pip install --upgrade pip
.\.venv-governed-repo\Scripts\python.exe -m pip install .\packages\governed-repo

For the remaining Windows commands, replace python with:

.\.venv-governed-repo\Scripts\python.exe

python -m venv .venv-governed-repo
./.venv-governed-repo/bin/python -m pip install --upgrade pip
./.venv-governed-repo/bin/python -m pip install ./packages/governed-repo

For the remaining macOS/Linux commands, replace python with:

./.venv-governed-repo/bin/python

Run:

python -c "import governed_repo; print(governed_repo.PromotionReason.ELIGIBLE_FOR_PROMOTION.value)"

Expected output:

ELIGIBLE_FOR_PROMOTION

Record whether this worked without changing the instructions.

Create a file named outside_operator_trial.py with this content:

from governed_repo import (
    ChangeIdentity,
    GateReceipt,
    GateRequirement,
    PromotionPolicy,
    assess_promotion,
)

identity = ChangeIdentity(
    repository="example/repository",
    change_id="trial-1",
    base_sha="base-123",
    head_sha="head-456",
    observed_base_sha="base-123",
    observed_head_sha="head-456",
    observed_at="2026-10-01T18:00:00Z",
)

policy = PromotionPolicy(
    required_gates=(
        GateRequirement(
            gate_id="tests",
            gate_class="quality",
            require_head_binding=True,
        ),
    ),
)

receipt = GateReceipt(
    gate_id="tests",
    status="completed",
    conclusion="success",
    observed_at="2026-10-01T18:00:00Z",
    head_sha="head-456",
)

result = assess_promotion(identity, policy, [receipt])
print(result.to_dict())

Run it:

python outside_operator_trial.py

The result should include:

eligible: true
reason_code: ELIGIBLE_FOR_PROMOTION
merge_executed: false
mutation_executed: false
authorization_effect: NONE

The exact Python dictionary formatting may differ, but those values should be present.

In the same file, change only:

observed_base_sha="older-base",

Run the file again.

The result should now include:

eligible: false
reason_code: STALE_BASE
merge_executed: false
mutation_executed: false
authorization_effect: NONE

Do not "fix" the denial. The expected denial is part of the test.

In your own words, answer:

1. What does ELIGIBLE_FOR_PROMOTION mean here?
2. Did Governed Repo merge, push, deploy, release, or mutate anything?
3. Does the receipt itself grant authorization to merge?
4. Why did the second case return STALE_BASE?
5. What information would you still need before taking a real repository action?

There is no requirement to use particular wording. The important boundary is that eligibility is understood as an evaluation result, not execution or an authorization grant.

Copy and complete this record:

Operator role/background (coarse):
Operating system:
Python version:
Trial date:
Exact source SHA:
Installation completed without undocumented steps: YES / NO
Import check passed: YES / NO
Eligible case produced ELIGIBLE_FOR_PROMOTION: YES / NO
Denial case produced STALE_BASE: YES / NO
Non-effect fields remained false/NONE: YES / NO

Did you need help beyond this document?:
If yes, what help?:

Commands or wording that were unclear:
Missing assumptions:
Unexpected errors:
Workarounds attempted:
What would you change in the instructions?:

Your explanation of ELIGIBLE_FOR_PROMOTION:
Your explanation of the non-effect fields:
Your explanation of STALE_BASE:

Overall trial completed without author coaching: YES / NO

Attach the completed record to issue #1210 or return it to the project owner without editing the answers after the fact.

This bounded trial is complete only if the operator can, without implementation-author coaching:

• install the package from the exact source;
• run one eligible case;
• run the expected stale-base denial;
• identify the reason codes;
• correctly understand that the receipt performs no mutation and grants no authorization.

A failed or confusing step should be recorded rather than hidden.

One successful trial can establish only:

bounded outside-operator usability for one operator, one documented package path, and the tested environment.

It does not establish:

• generalized usability;
• independent scientific validation;
• production security;
• complete GitHub correctness;
• merge authority;
• compliance or certification;
• High-Assurance authorization;
• publication readiness;
• commercial demand;
• product-market fit.