Add manifest-driven layered artifact generation command

[ProfRandom92]

Motivation

• Provide a single deterministic entrypoint to regenerate artifacts/layered_admissibility_results.json from fixtures/manifest.json using the existing manifest-driven generator while preserving exact rational scoring and byte-stable output.

Description

• Added scripts/generate_layered_admissibility_artifact.py which imports DegradationCurveGenerator, calls fixtures_for_layered_admissibility_curve(), generates the curve with curve_id = coding_workflow_pr_review_curve_v1, and writes artifacts/layered_admissibility_results.json via the generator's deterministic write_json behavior.
• Added tests/test_generate_layered_admissibility_artifact.py verifying the generated JSON equals the committed artifact, no new top-level fields are introduced, and the moderate fixture retains overall_admissibility_score = 0.8333333333333334.
• Added npm script "generate:layered-admissibility": "python scripts/generate_layered_admissibility_artifact.py" to package.json to provide a simple CLI shim.

Summary: Added a deterministic, manifest-driven artifact regeneration script, focused tests, and a small npm entry.

Changed files: scripts/generate_layered_admissibility_artifact.py, tests/test_generate_layered_admissibility_artifact.py, package.json.

Determinism guarantees: the script uses manifest ordering from fixtures/manifest.json, reuses DegradationCurveGenerator.write_json for stable serialization, and introduces no timestamps, hostnames, nondeterministic IDs, or environment-derived fields.

Testing

• Ran python scripts/generate_layered_admissibility_artifact.py which produced artifacts/layered_admissibility_results.json with the committed content.
• Ran unit tests used for validation: pytest tests/test_admissibility_scorer.py -q (10 passed), pytest tests/test_degradation_curve_generator.py -q (13 passed), pytest tests/test_fixture_manifest.py -q (8 passed), pytest tests/test_svg_curve_renderer.py -q (7 passed), and the full test suite via npm run check (all tests passed; full pytest run: 196 passed).

Validation commands run: python scripts/generate_layered_admissibility_artifact.py, pytest tests/test_admissibility_scorer.py -q, pytest tests/test_degradation_curve_generator.py -q, pytest tests/test_fixture_manifest.py -q, pytest tests/test_svg_curve_renderer.py -q, npm run check.

Risks: Low — this is a thin, deterministic wrapper around existing generator logic and focused regression tests; no core scoring, fixture, or schema changes were made.

Next: Open PR for review and keep the change small and focused; run CI to confirm cloud validation.

---

Codex Task

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
comptextv7	Ready	Preview, Comment	May 19, 2026 2:57pm

[gemini-code-assist]

Code Review

This pull request introduces a new script and associated tests for regenerating the layered admissibility artifact, along with a corresponding npm command. The feedback provided focuses on enhancing the robustness of path handling by anchoring file paths to the repository root in both the script and the test suite, ensuring consistent behavior across different execution environments.

[gemini-code-assist]

The OUTPUT_PATH is currently defined as a relative path. Since this script is intended to be a deterministic entrypoint and already calculates REPO_ROOT, it is more robust to anchor the output path to the repository root. This ensures the script behaves consistently regardless of the current working directory.

Suggested change

	OUTPUT_PATH = Path("artifacts/layered_admissibility_results.json")
	OUTPUT_PATH = REPO_ROOT / "artifacts" / "layered_admissibility_results.json"

[gemini-code-assist]

The call to fixtures_for_layered_admissibility_curve() relies on a default relative path for the manifest file. To ensure portability and consistency with the REPO_ROOT logic, it's better to pass an absolute path anchored to the repository root.

Suggested change

	fixtures = generator.fixtures_for_layered_admissibility_curve()
	fixtures = generator.fixtures_for_layered_admissibility_curve(manifest_path=REPO_ROOT / "fixtures" / "manifest.json")

[gemini-code-assist]

Import REPO_ROOT from the script to allow anchoring paths in the test file, ensuring tests can be run from any directory.

Suggested change

	from scripts.generate_layered_admissibility_artifact import (
	CURVE_ID,
	generate_layered_admissibility_artifact,
	)
	from scripts.generate_layered_admissibility_artifact import (
	CURVE_ID,
	REPO_ROOT,
	generate_layered_admissibility_artifact,
	)

[gemini-code-assist]

Anchor ARTIFACT_PATH to the repository root using the imported REPO_ROOT. This improves test robustness when executed from different working directories.

Suggested change

	ARTIFACT_PATH = Path("artifacts/layered_admissibility_results.json")
	ARTIFACT_PATH = REPO_ROOT / "artifacts" / "layered_admissibility_results.json"

[gemini-code-assist]

Pass the absolute manifest path to fixtures_for_layered_admissibility_curve to ensure the test can locate the manifest regardless of the current working directory.

Suggested change

	curve = generator.generate(generator.fixtures_for_layered_admissibility_curve(), curve_id=CURVE_ID)
	fixtures = generator.fixtures_for_layered_admissibility_curve(manifest_path=REPO_ROOT / "fixtures" / "manifest.json")
	curve = generator.generate(fixtures, curve_id=CURVE_ID)