# DecisionOps Test Project A simple Python project with intentionally flaky tests to demonstrate DecisionOps self-healing capabilities. ## How the flaky tests work | Test | Failure rate | Simulates | |---|---|---| | `test_flaky_network_call` | ~30% | Network timeout to upstream API | | `test_flaky_database_connection` | ~30% | Database connection pool exhaustion | | `test_flaky_external_service` | ~25% | Payment gateway rate limiting | | `test_flaky_cache_miss` | ~20% | Redis cache cleared mid-test | These are **transient failures** — running the exact same tests again will likely pass. This is what DecisionOps self-healing is designed for. ## Running locally ```bash pip install -r requirements.txt pytest test_app.py -v ``` Run multiple times and you will see different tests fail on each run. ## How DecisionOps healing works with this project 1. A push triggers GitHub Actions 2. A flaky test fails 3. GitHub Actions notifies DecisionOps via webhook 4. DecisionOps classifies the failure as `test_failure` (transient) 5. The self-healing engine retries the pipeline 6. The retry run passes (because the flaky condition resolved) 7. Healing tab shows ✓ Healed trigger