E2e Coverage
Testing & Qualitye2e-coverageE2e Coverage is an Agent Skill for testing & quality published in the meteor/meteor repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Use when adding, modifying, or reviewing E2E test apps/skeletons to keep the test coverage report up to date.
QUICK START
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/meteor/meteor/blob/HEAD/.github/skills/e2e-coverage/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/e2e-coverage, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
E2E Test Coverage Report
Guidelines for maintaining dev/modern-tools/rspack/E2E_COVERAGE.md — a single-page report of what every E2E app and skeleton tests.
When to Update
| Trigger | Action |
|---|---|
New app added to apps/ | Add a subsection under Apps with a coverage table |
New skeleton added to skeleton.test.js | Add a row to the Skeletons table |
| New npm package imported for compatibility testing | Add an entry under NPM Package Compatibility with the package name, file, and reason |
| New custom assertion added to a test file | Add a row to that app's coverage table |
| New feature tested across multiple apps | Add a row to the Feature Coverage Matrix |
| App or skeleton removed | Remove its entries from all sections |
Report Structure
The report has five sections, in this order:
- Test Lifecycle — the phases every app/skeleton goes through (init, run, prod, test, test once, build) and what default assertions apply
- Apps — one subsection per
apps/<name>/with a short description and a| What is covered | Phase |table - Skeletons — single table with one row per skeleton (
| Skeleton | Port | Language | Extra coverage |) - NPM Package Compatibility — grouped by app, each entry has the package name, file path, and why it's included (ESM-only, native bindings, subpath exports, etc.)
- Feature Coverage Matrix — cross-reference table (
| Feature | Apps | Skeletons |) showing where each capability is tested
How to Gather Information
For each app or skeleton, check these sources:
| Source | What to look for |
|---|---|
<name>.test.js | Test helper used, options (env, configFile, buildDir, testFullApp, checkBundleFilePaths), all customAssertions callbacks and what they assert |
skeleton.test.js | The testMeteorSkeleton({ skeletonName: '<name>' }) block for that skeleton |
apps/<name>/server/main.js | npm imports with comments explaining why (ESM-only, native bindings, etc.) |
apps/<name>/imports/ | Shared code with special imports (node: protocol, JSX packages) |
apps/<name>/rspack.config.* | Custom config features (compileWithRspack, compileWithMeteor, disablePlugins, custom rules) |
apps/<name>/package.json | Dependencies that exist solely for compatibility testing |
Writing Guidelines
- Keep descriptions short — one line per table row
- Use the phase names from the lifecycle table: Init, Run, Prod, Test, Build, All
- For npm packages, always state the reason (what module format issue it validates)
- Don't duplicate info between the per-app table and the feature matrix — the app table has detail, the matrix has the cross-reference
- When an env var is set in a test file, note it as
All (env prefix)in the phase column