Add Robot Framework E2E Tests for University Agent

[devin-ai-integration]

Description

This PR implements comprehensive end-to-end tests using Robot Framework for the University Agent, covering both the FastAPI REST API and Streamlit UI. The tests validate agent functionality including university searches, error handling, UI interactions, and service lifecycle management.

Link to Devin run: https://app.devin.ai/sessions/e663b31a138447dc9613ba3584e6cd55
Requested by: @thepollari

Changes

Test Infrastructure

• Created tests/e2e/robot/ directory structure with 7 new files
• api_tests.robot - 7 test cases covering API endpoints (health checks, country/name searches, error handling, timeouts)
• ui_tests.robot - 9 test cases covering Streamlit UI (page loading, button interactions, chat input, status checks)
• resources/keywords.robot - Reusable keywords for service management (start/stop agent & Streamlit, query sending, validation)
• resources/variables.robot - Common variables (URLs, timeouts, ports)
• fixtures/test_data.json - Test data with valid/invalid queries
• README.md - Comprehensive documentation on running tests, troubleshooting, and test coverage

Dependencies

• Added robotframework>=7.1.1, robotframework-browser>=19.10.1, robotframework-requests>=0.9.7 to pyproject.toml
• Updated uv.lock with new dependencies and transitive packages

CI/CD Integration

• Added robot-tests job to .github/workflows/check.yaml
• Job installs dependencies, initializes Browser Library, runs tests, and uploads test artifacts (log.html, report.html, output.xml)
• Requires GEMINI_API_KEY secret to be configured

Build Tools

• Added robot-tests target to Makefile for local test execution

Type of Change

• New feature
• Documentation

Test Coverage

API Tests (7 cases):

• Agent endpoint health check
• Search by country (Finland) with validation
• Search by name (Aalto University) with validation
• Empty query handling
• Response structure validation
• Connection error handling
• Timeout verification (<30s)

UI Tests (9 cases):

• Streamlit page loads at http://localhost:8502
• Page title displays correctly
• "Check Agent Status" button functionality
• All 5 example query buttons present
• Example button click functionality
• Chat input accepts queries
• Agent responses display in chat
• Clear conversation button
• Agent URL configuration in sidebar

Additional Comments

⚠️ Important for Reviewers

1. CI Secret Required - This PR assumes GEMINI_API_KEY is configured in GitHub Secrets. Please verify this secret exists before merging.

2. Not Fully Tested - Due to environment limitations (missing browser system dependencies locally), these tests were only validated via Robot Framework dry runs to verify syntax. Actual test execution has not been performed. The tests should work in CI but will need monitoring on first run.

3. Service Management - Tests use subprocess-based service lifecycle management (start agent on port 8001, Streamlit on port 8502). Hard-coded sleep times (10s for agent, 5s for Streamlit) may need adjustment based on CI performance.

4. External API Dependency - Tests use the real universities.hipolabs.com API rather than mocks (Task 7 was optional). Tests may fail if the external API is down.

5. Browser System Dependencies - The CI job runs rfbrowser init which requires system libraries. Standard Ubuntu GitHub Actions runners should have these, but this hasn't been verified.

Human Review Checklist

• Verify GEMINI_API_KEY secret is configured in GitHub repository settings
• Check if 10s agent startup wait and 5s Streamlit startup wait are sufficient
• Review service management approach (subprocess-based) for robustness
• Validate that all 8 tasks from original requirements are implemented
• Monitor first CI run to ensure tests actually pass
• Consider whether external API dependency is acceptable or if mocking should be added

Implementation Notes

All 8 tasks from the requirements were implemented:

1. ✅ Test infrastructure and directory structure
2. ✅ API test suite with 7 test cases
3. ✅ UI test suite with 9 test cases (updated to 9 based on actual implementation)
4. ✅ Reusable keywords for service management
5. ✅ Suite setup/teardown with environment configuration
6. ✅ Test data fixtures
7. ⚠️ External API mocking (framework created but marked optional - using real API)
8. ✅ CI/CD integration via GitHub Actions

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring