lee-dongwook/E2E Self-Heal
Run Playwright E2E tests and auto-heal broken selectors with an LLM.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| mode | Operating mode: heal (patch the test) or review (comment source-level fixes) | no | heal |
| test-path | Path to the Playwright test to heal/review when the suite fails | yes | — |
| nvidia-api-key | API key for the NVIDIA NIM LLM provider (E2E_HEALER_NVIDIA_API_KEY) | yes | — |
| diff-base | Base git ref to diff against as base...HEAD (e.g. the PR base sha) | no | "" |
| diff-file | Path to a diff file (relative to working-directory); takes precedence over diff-base | no | "" |
| working-directory | Directory to run Playwright and the healer from (where the Playwright project lives) | no | . |
| playwright-cmd | Command used to run Playwright | no | npx playwright test |
| app-url | URL the Selector Verifier loads to check patched selectors (empty = skip verification) | no | "" |
Outputs
| name | description |
|---|---|
| outcome | One of: passed | healed | unhealed | reviewed |
| summary-path | Path to the RepairSummary JSON (heal mode; empty otherwise) |
| review-path | Path to the ReviewReport JSON (review mode; empty otherwise) |