[lwaldron]

The eight fixtures from #36, rewritten to the Tests section of PROTOCOL_STANDARD.md (ADR 0016). Expected values are unchanged; only the shape is.

• Files are {path: ...}.
• The two orientation fixtures and invalid_parameters are diagnostic tests whose outcomes carry evidence: expected_outputs for a corrected run, expected_absent for a refusal.
• The compare_* and intended_* keys are gone; the checker rules replace them.

Validates locally against waldronlab/agent-protocol-standard#45. CI here uses the validator at main, which ignores tests/ until that merges, so merge that first.

Drafted with Claude Code.

[claude]

⚠️ Code review skipped — your organization has no extra usage available to pay for this review.

If your organization's extra usage balance is empty, an organization admin can add extra usage credits at claude.ai/admin-settings/usage. If its monthly spend limit was reached, an admin can raise it on the same page. If neither applies, contact Anthropic support.

Once extra usage is available, someone with write access to this repository can comment @claude review on this pull request to trigger a review.

[Copilot]

🟡 Changes recommended

The variance orientation diagnostic needs numeric tolerance to avoid rejecting correct results due to floating-point rounding.

1 open issue

• Set numeric tolerance for orientation-corrected variance checks · New

What changed in this PR

Migrates eight independent-filtering fixtures to the standard test schema.

Changes:

• Represents input and expected-output files as {path: ...}.
• Moves diagnostic expectations into accepted outcomes.
• Removes legacy checker flags and intended_* keys.

File	Description
protocols/​independent-filtering-variance/​tests/​variance_rows/​test.yaml	Migrates row-oriented variance expectations.
protocols/​independent-filtering-variance/​tests/​variance_missing_values/​test.yaml	Migrates missing-value variance expectations.
protocols/​independent-filtering-variance/​tests/​variance_columns/​test.yaml	Defines orientation correction and refusal outcomes.
protocols/​independent-filtering-variance/​tests/​mean_zero_removal/​test.yaml	Migrates zero-removal mean expectations.
protocols/​independent-filtering-variance/​tests/​mean_rows/​test.yaml	Migrates row-oriented mean expectations.
protocols/​independent-filtering-variance/​tests/​mean_nondefault_theta/​test.yaml	Migrates nondefault-threshold mean expectations.
protocols/​independent-filtering-variance/​tests/​mean_columns/​test.yaml	Defines orientation correction and refusal outcomes.
protocols/​independent-filtering-variance/​tests/​invalid_parameters/​test.yaml	Defines invalid-parameter refusal evidence.

_{Review effort: Balanced}

[Copilot]

🔵 Needs a closer look

The variance-columns migration drops three expected quantities that should be preserved in a shape-only update.

No open issues 🎉

1 previously missed issue

In code that hasn't changed since last review

Preserve expected output quantities in corrected outcome

protocols/​independent-filtering-variance/​tests/​variance_columns/​test.yaml:15

The migration removes intended_n_eligible: 5, intended_m: 2, and intended_cutoff: 1.3333333333333333 without preserving them under the corrected outcome. This loses expected quantities despite the PR's stated shape-only scope. The referenced standard compares only declared output keys, so a corrected run can omit or misreport these quantities and still pass. Preserve them as n_eligible, m, and cutoff in this outcome's expected_outputs.

1 resolved since last review

• Set numeric tolerance for orientation-corrected variance checks

_{Review effort: Balanced}