Repository navigation

feat: confidence-gap safeguards before wet-lab synthesis #26

	name: CI

	on:
	pull_request:
	push:
	branches: [main]

	jobs:
	test:
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4
	- uses: actions/setup-python@v5
	with:
	python-version: '3.11'

	- name: Install
	run: pip install -e ".[dev]"

	- name: Lint
	run: ruff check src tests

	- name: Unit + integration tests
	run: pytest -q --tb=short

	- name: Demo pipeline
	run: make demo

	- name: Validate evidence certificates
	run: \|
	for cert in outputs/evidence/*.json; do
	PYTHONPATH=src python -m openamp_foundry.cli validate \
	--certificate "$cert" \
	--schema schemas/candidate.schema.json
	done

	- name: Leakage check (informational)
	run: make bench-leakage

	- name: Hidden-active benchmark (require EF >= 1.5 at k=5)
	run: \|
	PYTHONPATH=src python -c "
	import json, subprocess, sys, os
	result = subprocess.run(
	['python', '-m', 'openamp_foundry.cli', 'bench', 'baseline',
	'--candidates', 'examples/benchmark/mixed_candidates.csv',
	'--positives', 'examples/benchmark/active_labels.csv',
	'--k', '5'],
	capture_output=True, text=True, check=True,
	env={**os.environ, 'PYTHONPATH': 'src'}
	)
	data = json.loads(result.stdout)
	ef = next(r['enrichment_factor'] for r in data['results'] if r['k'] == 5)
	print(f'Enrichment factor at k=5: {ef}')
	if ef < 1.5:
	print(f'FAIL: EF={ef} below minimum 1.5')
	sys.exit(1)
	print('PASS: Pipeline outperforms random ranker on hidden-active benchmark')
	"