chore: publish ArmaraOS CLI v0.8.4 for Intel macs (x86_64-apple-darwin) #656

Workflow file for this run

	name: CI

	on:
	push:
	branches: [main, master]
	pull_request:
	branches: [main, master]
	schedule:
	- cron: "0 3 * * *"
	workflow_dispatch:

	jobs:
	docs-contract:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4
	with:
	fetch-depth: 0

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.10"

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev]

	- name: Run docs contract check (PR diff-aware)
	if: github.event_name == 'pull_request'
	run: \|
	python scripts/check_docs_contracts.py --scope changed --base-ref "origin/${{ github.base_ref }}"

	- name: Run docs contract check (full)
	if: github.event_name != 'pull_request'
	run: \|
	python scripts/check_docs_contracts.py --scope all

	core-pr:
	if: github.event_name != 'schedule'
	runs-on: ${{ matrix.os }}
	strategy:
	fail-fast: false
	matrix:
	os: [ubuntu-latest, macos-latest, windows-latest]
	python-version: ["3.10", "3.11"]
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev]

	- name: Run core profile
	run: python scripts/run_test_profiles.py --profile core

	# Named gate: a2a unit tests + local threaded HTTPServer integration (skips bind in strict sandboxes).
	- name: A2A adapter (tests.test_a2a_adapter + tests.test_a2a_adapter_integration)
	run: python -m pytest -q tests/test_a2a_adapter.py tests/test_a2a_adapter_integration.py

	# Explicit gate: compile → RuntimeEngine → SolanaAdapter (includes dry-run mutating verbs).
	# Core profile already runs this file; this step names the contract in CI logs.
	- name: Solana prediction-market E2E (dry-run runtime)
	run: python -m pytest -q tests/test_solana_prediction_market_e2e.py

	- name: Verify adapter manifest consistency
	run: python -m pytest -q tests/test_adapter_manifest.py

	- name: Verify artifact profile consistency
	run: \|
	python scripts/check_all_strict.py
	python scripts/check_all_nonstrict.py
	python -m pytest -q tests/test_artifact_profiles.py

	parser-compat:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.10"

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev]

	- name: Validate S+cron schedule lines (IR services.path)
	run: python scripts/validate_s_cron_schedules.py

	- name: Run parser compatibility gate
	run: \|
	python -m pytest -q \
	tests/test_compiler_patch_a_guardrails.py \
	tests/test_sanity_fixes.py \
	tests/test_ir_canonical.py \
	tests/test_emits_artifacts.py \
	tests/test_conformance.py::test_minimal_api_emit_openapi_has_path

	- name: Wishlist examples strict + no-network smoke (01, 05b)
	run: \|
	set -euo pipefail
	python -m pytest -q tests/test_wishlist_examples_strict.py
	python -m cli.main run examples/wishlist/01_cache_and_memory.ainl --json \
	--frame-json '{"session_key":"ci","note":"hello"}' 2>/dev/null \
	\| python -c "import sys,json; d=json.load(sys.stdin); assert d.get('ok') is True, d"
	python -m cli.main run examples/wishlist/05b_unified_llm_offline_config.ainl --json \
	--config examples/wishlist/fixtures/llm_offline.yaml \
	--frame-json '{"user_query":"ci rust smoke"}' 2>/dev/null \
	\| python -c "import sys,json; d=json.load(sys.stdin); assert d.get('ok') is True, d; assert 'OFFLINE_LLM' in str(d.get('result')), d"

	# No outbound network required: a2a unit tests only (integration may skip in stricter sandboxes; core-pr also runs both).
	a2a-adapter-unit:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.10"

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev]

	- name: A2A adapter unit tests only
	run: python -m pytest -q tests/test_a2a_adapter.py

	conformance-matrix:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.10"

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev]

	- name: Run conformance matrix
	run: make conformance

	openapi-compat:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.10"

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev]

	- name: Run OpenAPI compatibility gate
	run: \|
	python -m pytest -q \
	tests/test_compiler_patch_c_openapi_types.py \
	tests/test_conformance.py::test_minimal_api_emit_openapi_has_path \
	tests/test_lossless.py::test_openapi_array_inner_type_ref_for_model \
	tests/test_lossless.py::test_openapi_json_type_maps_to_object_schema \
	tests/test_lossless.py::test_openapi_response_data_allows_array_or_object

	apollo-promoter:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.10"

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev]

	- name: Apollo promoter safety net (strict + gateway tests)
	run: make check-apollo-promoter

	compiler-compat-required:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	needs: [parser-compat, openapi-compat]
	steps:
	- name: Compatibility gates passed
	run: echo "parser-compat and openapi-compat both passed"

	integration-nightly:
	if: github.event_name == 'schedule' \|\| github.event_name == 'workflow_dispatch'
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.10"

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev]

	- name: Run full profile
	run: python scripts/run_test_profiles.py --profile full

	# Benchmarks + JSON regression gate: prefers committed CI baselines (tooling/benchmark_*_ci.json)
	# when present on the base SHA; otherwise falls back to full reports (benchmark_size.json,
	# benchmark_runtime_results.json).
	benchmark-regression:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	env:
	BENCH_THRESHOLD: "0.10"
	steps:
	- uses: actions/checkout@v4
	with:
	fetch-depth: 0

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.10"

	- name: Cache pip
	uses: actions/cache@v4
	with:
	path: ~/.cache/pip
	key: ${{ runner.os }}-pip-bench-${{ hashFiles('pyproject.toml') }}
	restore-keys: \|
	${{ runner.os }}-pip-bench-

	- name: Install deps (dev + benchmark)
	run: \|
	python -m pip install --upgrade pip
	pip install .[dev,benchmark]

	- name: Resolve baseline SHA
	id: baseline
	run: \|
	git fetch origin main --depth=1 2>/dev/null \|\| true
	if [ "${{ github.event_name }}" = "pull_request" ]; then
	echo "sha=${{ github.event.pull_request.base.sha }}" >> "$GITHUB_OUTPUT"
	elif [ "${{ github.event_name }}" = "push" ] \
	&& [ -n "${{ github.event.before }}" ] \
	&& [ "${{ github.event.before }}" != "0000000000000000000000000000000000000000" ]; then
	echo "sha=${{ github.event.before }}" >> "$GITHUB_OUTPUT"
	else
	echo "sha=$(git rev-parse origin/main 2>/dev/null \|\| echo "")" >> "$GITHUB_OUTPUT"
	fi

	- name: Extract baseline JSON (from git)
	id: baselines
	run: \|
	set -euo pipefail
	SHA="${{ steps.baseline.outputs.sha }}"
	mkdir -p /tmp/bench-baseline
	if [ -z "$SHA" ]; then
	echo "size_ci=missing" >> "$GITHUB_OUTPUT"
	echo "size=missing" >> "$GITHUB_OUTPUT"
	echo "runtime_ci=missing" >> "$GITHUB_OUTPUT"
	echo "runtime=missing" >> "$GITHUB_OUTPUT"
	exit 0
	fi
	if git cat-file -e "${SHA}:tooling/benchmark_size_ci.json" 2>/dev/null; then
	git show "${SHA}:tooling/benchmark_size_ci.json" > /tmp/bench-baseline/benchmark_size_ci.json
	echo "size_ci=present" >> "$GITHUB_OUTPUT"
	else
	echo "size_ci=missing" >> "$GITHUB_OUTPUT"
	fi
	if git cat-file -e "${SHA}:tooling/benchmark_size.json" 2>/dev/null; then
	git show "${SHA}:tooling/benchmark_size.json" > /tmp/bench-baseline/benchmark_size.json
	echo "size=present" >> "$GITHUB_OUTPUT"
	else
	echo "size=missing" >> "$GITHUB_OUTPUT"
	fi
	if git cat-file -e "${SHA}:tooling/benchmark_runtime_ci.json" 2>/dev/null; then
	git show "${SHA}:tooling/benchmark_runtime_ci.json" > /tmp/bench-baseline/benchmark_runtime_ci.json
	echo "runtime_ci=present" >> "$GITHUB_OUTPUT"
	else
	echo "runtime_ci=missing" >> "$GITHUB_OUTPUT"
	fi
	if git cat-file -e "${SHA}:tooling/benchmark_runtime_results.json" 2>/dev/null; then
	git show "${SHA}:tooling/benchmark_runtime_results.json" > /tmp/bench-baseline/benchmark_runtime_results.json
	echo "runtime=present" >> "$GITHUB_OUTPUT"
	else
	echo "runtime=missing" >> "$GITHUB_OUTPUT"
	fi

	- name: Run size benchmark (CI slice)
	run: \|
	python scripts/benchmark_size.py \
	--mode minimal_emit \
	--profile-name canonical_strict_valid \
	--compare-baselines \
	--cost-model gpt-4o \
	--json-out tooling/benchmark_size_ci.json \
	--skip-markdown-inject

	- name: Run runtime benchmark
	run: \|
	python scripts/benchmark_runtime.py \
	--profile-name canonical_strict_valid \
	--compare-baselines \
	--reliability-runs 10 \
	--cost-model gpt-4o \
	--json-out tooling/benchmark_runtime_ci.json \
	--skip-markdown-inject

	- name: Upload benchmark JSON (pre-regression)
	uses: actions/upload-artifact@v4
	with:
	name: benchmark-json-precheck-${{ github.run_id }}
	if-no-files-found: warn
	path: \|
	tooling/benchmark_size_ci.json
	tooling/benchmark_runtime_ci.json

	- name: Regression check (size)
	if: steps.baselines.outputs.size_ci == 'present' \|\| steps.baselines.outputs.size == 'present'
	run: \|
	set -euo pipefail
	if [ "${{ steps.baselines.outputs.size_ci }}" = "present" ]; then
	OLD_JSON=/tmp/bench-baseline/benchmark_size_ci.json
	else
	OLD_JSON=/tmp/bench-baseline/benchmark_size.json
	fi
	python scripts/compare_benchmark_json.py \
	--old-json "$OLD_JSON" \
	--new-json tooling/benchmark_size_ci.json \
	--threshold "$BENCH_THRESHOLD"

	- name: Regression check (size baseline missing)
	if: steps.baselines.outputs.size_ci != 'present' && steps.baselines.outputs.size != 'present'
	run: \|
	echo "::notice::No tooling/benchmark_size_ci.json or benchmark_size.json at baseline SHA; skipping size regression compare."

	- name: Regression check (runtime)
	id: runtime_regression
	if: (steps.baselines.outputs.runtime_ci == 'present' \|\| steps.baselines.outputs.runtime == 'present') && github.event_name != 'pull_request'
	continue-on-error: true
	run: \|
	set -euo pipefail
	if [ "${{ steps.baselines.outputs.runtime_ci }}" = "present" ]; then
	OLD_JSON=/tmp/bench-baseline/benchmark_runtime_ci.json
	else
	OLD_JSON=/tmp/bench-baseline/benchmark_runtime_results.json
	fi
	python scripts/compare_benchmark_json.py \
	--old-json "$OLD_JSON" \
	--new-json tooling/benchmark_runtime_ci.json \
	--threshold "$BENCH_THRESHOLD"

	- name: Runtime benchmark compare (PR informational)
	if: (steps.baselines.outputs.runtime_ci == 'present' \|\| steps.baselines.outputs.runtime == 'present') && github.event_name == 'pull_request'
	run: \|
	set -euo pipefail
	if [ "${{ steps.baselines.outputs.runtime_ci }}" = "present" ]; then
	OLD_JSON=/tmp/bench-baseline/benchmark_runtime_ci.json
	else
	OLD_JSON=/tmp/bench-baseline/benchmark_runtime_results.json
	fi
	python scripts/compare_benchmark_json.py \
	--old-json "$OLD_JSON" \
	--new-json tooling/benchmark_runtime_ci.json \
	--threshold "$BENCH_THRESHOLD" \|\| true

	- name: Runtime baseline missing (informational)
	if: steps.baselines.outputs.runtime_ci != 'present' && steps.baselines.outputs.runtime != 'present'
	run: \|
	echo "::notice::No tooling/benchmark_runtime_ci.json or benchmark_runtime_results.json at baseline SHA; commit those files on main to enable runtime regression checks."

	- name: Upload benchmark JSON artifacts
	uses: actions/upload-artifact@v4
	with:
	name: benchmark-json-${{ github.run_id }}
	if-no-files-found: warn
	path: \|
	tooling/benchmark_size_ci.json
	tooling/benchmark_runtime_ci.json

	install-smoke:
	if: github.event_name != 'schedule'
	runs-on: ${{ matrix.os }}
	strategy:
	fail-fast: false
	matrix:
	os: [ubuntu-latest, macos-latest]
	python-version: ["3.10", "3.11", "3.12", "3.13"]
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}

	- name: Normal install smoke (venv)
	shell: bash
	run: \|
	set -euo pipefail
	python -m venv .venv-smoke
	. .venv-smoke/bin/activate
	python -m pip install --upgrade pip
	python -m pip install ".[mcp]"
	python -m pip check
	ainl --help >/dev/null
	ainl-mcp --help >/dev/null

	- name: User install smoke (no sudo)
	shell: bash
	run: \|
	set -euo pipefail
	export PYTHONUSERBASE="${RUNNER_TEMP}/ainl-userbase"
	mkdir -p "${PYTHONUSERBASE}"
	python -m pip install --upgrade pip
	python -m pip install --user --constraint constraints/py313-mcp.txt ".[mcp]"
	python -m pip check
	"${PYTHONUSERBASE}/bin/ainl" --help >/dev/null
	"${PYTHONUSERBASE}/bin/ainl-mcp" --help >/dev/null

	- name: Break-system-packages install smoke
	shell: bash
	run: \|
	set -euo pipefail
	python -m pip install --upgrade pip
	python -m pip install --break-system-packages --constraint constraints/py313-mcp.txt ".[mcp]"
	python -m pip check
	SCRIPT_DIR="$(python -c 'import sysconfig; print(sysconfig.get_path("scripts"))')"
	"${SCRIPT_DIR}/ainl" --help >/dev/null
	"${SCRIPT_DIR}/ainl-mcp" --help >/dev/null

	wheel-integrity:
	if: github.event_name != 'schedule'
	runs-on: ${{ matrix.os }}
	strategy:
	fail-fast: false
	matrix:
	os: [ubuntu-latest, macos-latest]
	python-version: ["3.10", "3.11", "3.12", "3.13"]
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}

	- name: Build wheel
	shell: bash
	run: \|
	set -euo pipefail
	python -m pip install --upgrade pip build
	python -m build

	- name: Wheel install integrity + imports
	shell: bash
	run: \|
	set -euo pipefail
	python -m venv .venv-wheel
	. .venv-wheel/bin/activate
	python -m pip install --upgrade pip
	WHEEL="$(ls dist/*.whl \| head -n 1)"
	python -m pip install --force-reinstall --constraint constraints/py313-mcp.txt "${WHEEL}[mcp]"
	python -m pip check
	python -c "import runtime.compat, adapters, cli.main"
	ainl --help >/dev/null
	ainl-mcp --help >/dev/null

	container-install-smoke:
	if: github.event_name != 'schedule'
	runs-on: ubuntu-latest
	strategy:
	fail-fast: false
	matrix:
	image: ["python:3.10-slim", "python:3.11-slim", "python:3.12-slim", "python:3.13-slim"]
	container:
	image: ${{ matrix.image }}
	steps:
	- uses: actions/checkout@v4

	- name: Non-root container install smoke
	shell: bash
	run: \|
	set -euo pipefail
	useradd -m -u 1000 ainl 2>/dev/null \|\| true
	chown -R 1000:1000 "$GITHUB_WORKSPACE"
	su -s /bin/bash ainl -c '
	cd "$GITHUB_WORKSPACE" && \
	python -m venv .venv-container && \
	. .venv-container/bin/activate && \
	python -m pip install --upgrade pip && \
	python -m pip install --constraint constraints/py313-mcp.txt ".[mcp]" && \
	python -m pip check && \
	ainl --help >/dev/null && \
	ainl-mcp --help >/dev/null
	'

Navigation Menu

chore: publish ArmaraOS CLI v0.8.4 for Intel macs (x86_64-apple-darwin) #656

Workflow file

chore: publish ArmaraOS CLI v0.8.4 for Intel macs (x86_64-apple-darwin) #656

Uh oh!

Workflow file for this run