docs: comprehensive README — metrics table, real capstone-rag results… #18

	name: AgentScope CI

	on:
	push:
	branches: [main]
	pull_request:
	branches: [main]

	jobs:
	test:
	name: Metric regression gates
	runs-on: ubuntu-latest

	steps:
	- name: Checkout
	uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.11"

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

	- name: Install dependencies
	run: pip install -r requirements.txt

	- name: Run metric regression gates
	run: \|
	python -m pytest \
	tests/test_ir_evaluator.py \
	tests/test_agent_behavior.py \
	tests/test_cost_analyzer.py \
	tests/test_geval_tool.py \
	tests/test_adversarial_eval.py \
	tests/test_synth_gen.py \
	tests/test_compiler.py \
	--tb=short -q
	env:
	# API keys are not needed for these tests — all LLM calls are mocked
	ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
	OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}

Navigation Menu