Integration test

.github/copilot-instructions.md

@@ -0,0 +1,30 @@
+# Copilot Instructions for Immermatch
+
+## Environment
+
+- **Always** activate the virtual environment first: `source .venv/bin/activate`
+- Python 3.10+, all dependencies installed in `.venv`
+- Gemini model: `gemini-3-flash-preview` via `google-genai` package (NOT the deprecated `google.generativeai`)
+
+## After every code change
+
+Run the full check suite without asking — just do it:
+
+```bash
+source .venv/bin/activate && pytest tests/ -x -q && ruff check . && mypy .
+```
+
+## Testing conventions
+
+- **Framework:** pytest + pytest-cov
+- **Test file naming:** `tests/test_<module>.py` for `immermatch/<module>.py`
+- **Mock all external services** (Gemini API, SerpAPI, Supabase, Resend) — no API keys needed to run tests
+- **Shared fixtures** in `tests/conftest.py`: `sample_profile`, `sample_job`, `sample_evaluation`, `sample_evaluated_job`
+- **Test fixture files** (sample CVs) live in `tests/fixtures/`
+- Pydantic models live in `immermatch/models.py` — follow existing patterns
+
+## Code conventions
+
+- All DB writes use `get_admin_client()`, never the anon client
+- Log subscriber UUIDs, never email addresses
+- All `st.error()` calls show generic messages; real exceptions go to `logger.exception()`

.github/workflows/ci.yml

@@ -46,7 +46,7 @@ jobs:
           cache: pip
       - run: pip install -e ".[test]"
       - name: Tests
-        run: pytest -v --cov=immermatch --cov-report=term
+        run: pytest -v --cov=immermatch --cov-report=term --cov-fail-under=50
 
   audit:
     runs-on: ubuntu-latest

.gitignore

@@ -11,6 +11,7 @@ __pycache__/
 *$py.class
 *.so
 .Python
+.coverage
 build/
 develop-eggs/
 dist/
@@ -41,7 +42,7 @@ Thumbs.db
 # Project specific
 *.pdf
 output/
-*cv*.md
+total_cv.md
 .immermatch_cache/
 
 # Streamlit secrets (contains API keys)

.pre-commit-config.yaml

@@ -41,9 +41,7 @@ repos:
     hooks:
       - id: test-suite
         name: Run tests
-        entry: pytest tests/ -x -q --tb=short
-        language: python
-        additional_dependencies:
-          - pytest
+        entry: .venv/bin/pytest tests/ -x -q --tb=short
+        language: system
         pass_filenames: false
         stages: [pre-push]

.secrets.baseline

@@ -90,6 +90,10 @@
     {
       "path": "detect_secrets.filters.allowlist.is_line_allowlisted"
     },
+    {
+      "path": "detect_secrets.filters.common.is_baseline_file",
+      "filename": ".secrets.baseline"
+    },
     {
       "path": "detect_secrets.filters.common.is_ignored_due_to_verification_policies",
       "min_level": 2
@@ -123,5 +127,5 @@
     }
   ],
   "results": {},
-  "generated_at": "2026-02-21T16:02:22Z"
+  "generated_at": "2026-02-28T18:42:35Z"
 }

AGENTS.md

@@ -544,6 +544,12 @@ source .venv/bin/activate
 # All:     ruff check . && mypy . && pytest tests/ -x -q
 ```
 
+**IMPORTANT:** After every code change, run the check suite **without asking for permission** — just do it:
+```bash
+source .venv/bin/activate && pytest tests/ -x -q && ruff check . && mypy .
+```
+Do not ask the user "Shall I run the tests?" — always run them automatically.
+
 ### Conventions for AI agents
 
 - **Always activate the virtual environment** (`source .venv/bin/activate`) before running any command (`pytest`, `ruff`, `mypy`, `streamlit`, etc.). The project's dependencies are installed only in `.venv`.

ROADMAP.md

@@ -8,7 +8,7 @@ Based on the current state (private repo, hosted on Streamlit Community Cloud) a
 
 ### 0.1 — Testing & CI
 - [x] **Write unit tests** for core modules: `llm.py`, `cache.py`, `cv_parser.py`, `db.py`, `search_agent.py`, `evaluator_agent.py` (mock API calls) — 146 tests across 9 test files
-- [ ] **Write integration tests** for the full pipeline (profile → queries → search → evaluate → summary) using fixture CVs
+- [x] **Write integration tests** for the full pipeline (profile → queries → search → evaluate → summary) using fixture CVs — 11 tests in `tests/test_integration.py` with tech + sustainability CV fixtures
 - [x] **Set up GitHub Actions CI** — run `pytest` on every push/PR, lint with `ruff`
 - [x] **Add type checking** — run `mypy` in CI (Pydantic models already help here)
 

pyproject.toml

@@ -51,6 +51,7 @@ include = ["immermatch*"]
 
 [tool.pytest.ini_options]
 testpaths = ["tests"]
+markers = ["integration: Full pipeline integration tests (profile → queries → search → evaluate → summary)"]
 
 [tool.ruff]
 target-version = "py310"
@@ -64,6 +65,20 @@ ignore = ["E501"]
 "daily_task.py" = ["E402"]
 "tests/**" = ["S101", "S105", "S106"]
 
+[tool.coverage.run]
+source = ["immermatch", "daily_task"]
+omit = [
+    "immermatch/pages/impressum.py",
+    "immermatch/pages/privacy.py",
+]
+
+[tool.coverage.report]
+exclude_lines = [
+    "pragma: no cover",
+    "if __name__ == .__main__.",
+    "if TYPE_CHECKING:",
+]
+
 [tool.mypy]
 ignore_missing_imports = true
 warn_unused_configs = true

tests/fixtures/sustainability_cv.md

@@ -0,0 +1,46 @@
+# Areeba Ilyas Qureshi
+## Climate & Sustainability Strategist | Environmental Engineer
+
+**Location:** Munich, Germany
+
+---
+
+### Professional Summary
+Results-driven Environmental Engineer and Sustainability Strategist with over 3 years of experience guiding global corporations toward Net Zero. Expert in GHG Protocol, CSRD/ESRS, SBTi, and ISO 14001. Proven track record in carbon accounting (Scopes 1-3), Life Cycle Assessments, and stakeholder engagement.
+
+---
+
+### Work Experience
+
+**Sustainability Consultant** at Global Climate GmbH (May 2023 – Present)
+- Key account management for 10-15 global manufacturing corporations
+- Authored 25+ sustainability reports (CSRD, ESRS, GRI, TCFD)
+- Developed Net Zero roadmaps and SBTi targets
+- Conducted 20+ GHG assessments (CCF, PCF, Scopes 1-3)
+
+**Research Assistant** at Technical University of Munich (May 2021 – April 2023)
+- Designed circular economy wastewater management concepts
+- Developed environmental models using MATLAB and Python
+- Supported 150+ international sustainability projects
+
+**VP Customer Experience** at AIESEC Munich (Feb 2020 – Feb 2021)
+- Led international team of 10, managed CRM with HubSpot
+- Built task management tools using Excel
+
+**Intern – Construction Management** at NESPAK (Oct 2017 – Nov 2017)
+- Urban planning research and infrastructure design using AutoCAD
+
+---
+
+### Education
+- MSc Environmental Engineering, Technical University of Munich (2019–2023)
+- BEng Urban and Infrastructure Planning, NED University (2014–2018)
+
+### Skills
+CSRD, ESRS, GRI, TCFD, SBTi, GHG Protocol, ISO 14001, ISO 14064, LCA, Carbon Accounting, MATLAB, Python, Power BI, AutoCAD, ArcGIS, HubSpot, Agile Project Management, Stakeholder Management
+
+### Languages
+- Urdu: Native
+- English: C1/C2
+- German: B1/B2
+- Hindi: Spoken

tests/test_app_ui.py

@@ -0,0 +1,60 @@
+"""Tests for the Streamlit main app UI (app.py).
+
+Uses Streamlit's AppTest framework to verify Phase A landing page renders,
+the consent checkbox is present, and sidebar elements appear.
+"""
+
+from unittest.mock import MagicMock, patch
+
+from streamlit.testing.v1 import AppTest
+
+_FAKE_ENV = {
+    "GOOGLE_API_KEY": "fake-google-key",  # pragma: allowlist secret
+    "SERPAPI_KEY": "fake-serpapi-key",  # pragma: allowlist secret
+    "SUPABASE_URL": "https://fake.supabase.co",
+    "SUPABASE_KEY": "fake-anon-key",  # pragma: allowlist secret
+    "SUPABASE_SERVICE_KEY": "fake-service-key",  # pragma: allowlist secret
+    "RESEND_API_KEY": "fake-resend-key",  # pragma: allowlist secret
+    "RESEND_FROM": "test@example.com",
+    "APP_URL": "https://app.example.com",
+}
+
+APP_FILE = "immermatch/app.py"
+
+
+class TestPhaseALanding:
+    """Phase A: no CV uploaded — landing page should render."""
+
+    @patch.dict("os.environ", _FAKE_ENV, clear=False)
+    @patch("immermatch.db.get_admin_client", return_value=MagicMock())
+    @patch("immermatch.db.purge_inactive_subscribers", return_value=0)
+    def test_app_loads_without_errors(self, _mock_purge: MagicMock, _mock_db: MagicMock) -> None:
+        at = AppTest.from_file(APP_FILE)
+        at.run()
+
+        # The app should not raise any uncaught exceptions
+        assert not at.exception, f"App raised exception: {at.exception}"
+
+    @patch.dict("os.environ", _FAKE_ENV, clear=False)
+    @patch("immermatch.db.get_admin_client", return_value=MagicMock())
+    @patch("immermatch.db.purge_inactive_subscribers", return_value=0)
+    def test_consent_checkbox_present(self, _mock_purge: MagicMock, _mock_db: MagicMock) -> None:
+        at = AppTest.from_file(APP_FILE)
+        at.run()
+
+        # GDPR consent checkbox should be on the page
+        checkboxes = at.checkbox
+        consent_found = any(
+            "consent" in (cb.label or "").lower() or "agree" in (cb.label or "").lower() for cb in checkboxes
+        )
+        assert consent_found, f"No consent checkbox found. Checkboxes: {[cb.label for cb in checkboxes]}"
+
+    @patch.dict("os.environ", _FAKE_ENV, clear=False)
+    @patch("immermatch.db.get_admin_client", return_value=MagicMock())
+    @patch("immermatch.db.purge_inactive_subscribers", return_value=0)
+    def test_sidebar_renders(self, _mock_purge: MagicMock, _mock_db: MagicMock) -> None:
+        at = AppTest.from_file(APP_FILE)
+        at.run()
+
+        # Sidebar should have a slider (min score)
+        assert len(at.slider) >= 1, "Expected at least one slider (min score) in sidebar"