agentleFS
MicrosoftKnown publishermicrosoft/vscode-java-dependency/.github/skills/uitest/SKILL.md

uitest

Skill199 starsChanged 4 months agoMITView on GitHub

Write, update, run, or debug vscode-java-dependency (Project Manager for Java) UI/E2E tests using AutoTest YAML plans. Use when the user asks for a UI test, E2E test, VS Code UI validation, Java Projects tree/view test, referenced-library test, or autotest plan.

  • Installs packages

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 1

    1. microsoft/vscode-java-pack★ 35538% same

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 4 are distinctive, and 10 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 10

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this skill to add or update UI/E2E coverage for vscode-java-dependency (Project Manager for Java).

    line 8
  2. Very specific:

    Update the existing plan when possible.

    line 21
  3. Distinctive:

    Use stable AutoTest actions and deterministic verifiers.

    line 22
  4. Distinctive:

    If validating the current branch, build the OSGi bundle and package the extension:

    line 29
  5. Distinctive:

    Run the plan against the packaged VSIX:

    line 37
  6. Very specific:

    Add --no-llm to skip natural-language verify: fields and rely only on deterministic verifiers for a fast local loop.

    line 43
  7. Very specific:

    Product code fix: after editing extension source (src/) or the OSGi bundle (jdtls.ext/), re-run step 5 (rebuild + repackage the VSIX) …

    line 48
  8. Very specific:

    Use --vsix vscode-java-dependency.vsix to test current-branch changes.

    line 65
  9. Very specific:

    Use ../invisible (not ../maven) for referenced-library / classpath commands, which only apply to unmanaged-folder projects.

    line 66
  10. Very specific:

    Prefer executeVSCodeCommand <commandId> for command-driven UI (e.g.

    line 67
  11. Very specific:

    Prefer verifyTreeItem for tree state, verifyFile for generated/modified/deleted files, verifyEditorTab for opened tabs, and verifyClipboard …

    line 69
  12. Very specific:

    Use waitForLanguageServer before tree interactions, and insertLineInFile for Java source edits that JDT LS must observe.

    line 70
  13. Very specific:

    Keep step IDs unique, descriptive, and kebab-case.

    line 72
  14. Distinctive:

    Avoid hard-coded coordinates and brittle DOM structure assumptions.

    line 73

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

UI/E2E tests with AutoTest

Use this skill to add or update UI/E2E coverage for vscode-java-dependency (Project Manager for Java).

The repository uses @vscjava/vscode-autotest: YAML plans in test/e2e-plans/*.yaml launch VS Code, install the Extension Pack for Java (vscjava.vscode-java-pack) plus a local VSIX of this extension, execute user-facing actions against the Java Projects view, capture screenshots, and write test-results/<plan>/results.json.

Prerequisites (local)

  • Node.js >= 18 and JDK 21+ installed and on PATH (JDK 21 is required to build the jdtls.ext OSGi bundle).
  • Close any running VS Code instance before running a plan locally; a running instance can block AutoTest from launching its own VS Code.
  • Workspace fixtures are in-repo — no external clones are needed. Plans reference ../maven (test/maven, a maven-archetype-quickstart project) or ../invisible (test/invisible, an unmanaged-folder project).

Workflow

  1. Identify the scenario and search test/e2e-plans/*.yaml for an existing plan that already covers the area (project explorer, view modes, classpath, export jar, new types, file operations, delete, copy paths, refresh, build lifecycle, autorefresh).
  2. Update the existing plan when possible. Create a new test/e2e-plans/java-dep-<scenario>.yaml only when no existing plan fits.
  3. Use stable AutoTest actions and deterministic verifiers. Do not add raw Playwright tests or screenshot-only checks.
  4. Validate the plan:
npx -y @vscjava/vscode-autotest validate test\e2e-plans\<name>.yaml
  1. If validating the current branch, build the OSGi bundle and package the extension:
npm install          # first time only; on later iterations run just the two commands below
npm run build-server
npx @vscode/vsce package -o vscode-java-dependency.vsix
  1. Run the plan against the packaged VSIX:
npx -y @vscjava/vscode-autotest run test\e2e-plans\<name>.yaml --vsix vscode-java-dependency.vsix --output test-results\<name>

Add --no-llm to skip natural-language verify: fields and rely only on deterministic verifiers for a fast local loop. Run the whole suite with npm run test-e2e (autotest run-all test/e2e-plans --no-llm).

  1. Inspect test-results/<name>/results.json and test-results/<name>/screenshots/.
  2. Iterate based on the failure cause:
    • Incorrect plan: fix the YAML and rerun step 6. No rebuild is needed.
    • Product code fix: after editing extension source (src/**) or the OSGi bundle (jdtls.ext/**), re-run step 5 (rebuild + repackage the VSIX) before rerunning step 6. Never rerun against a stale VSIX.
    • Product bug (report only): report the observed behavior and cite the failing step, screenshot, and result reason.

Authoring rules

  • For most plans, use:
setup:
  extension: "vscjava.vscode-java-pack"
  vscodeVersion: "stable"
  workspace: "../maven"
  settings:
    java.configuration.checkProjectSettingsExclusions: false
    workbench.startupEditor: "none"
  • Use --vsix vscode-java-dependency.vsix to test current-branch changes; do not rely on a marketplace copy of vscjava.vscode-java-dependency.
  • Use ../invisible (not ../maven) for referenced-library / classpath commands, which only apply to unmanaged-folder projects.
  • Prefer executeVSCodeCommand <commandId> for command-driven UI (e.g. javaProjectExplorer.focus, java.view.package.revealInProjectExplorer, workbench.actions.treeView.javaProjectExplorer.collapseAll).
  • Drive the tree with expandTreeItem <name> and title-bar buttons with clickViewTitleAction "Java Projects" "<action>".
  • Prefer verifyTreeItem for tree state, verifyFile for generated/modified/deleted files, verifyEditorTab for opened tabs, and verifyClipboard for copy-path commands.
  • Use waitForLanguageServer before tree interactions, and insertLineInFile for Java source edits that JDT LS must observe.
  • Free sidebar space (closeAuxiliaryBar, collapseSidebarSection OUTLINE/TIMELINE, collapseWorkspaceRoot) before asserting tree rows.
  • Keep step IDs unique, descriptive, and kebab-case. Omit verify: on steps whose only assertion is a deterministic verifier.
  • Avoid hard-coded coordinates and brittle DOM structure assumptions.

CI

The repository workflow .github/workflows/e2eUI.yml runs on push and pull requests to main. It lints, discovers test/e2e-plans/*.yaml into a matrix, builds a branch VSIX per OS, runs every plan on Windows and Linux as independent matrix cells, and uploads test-results/ artifacts plus an aggregate summary.

Each plan surfaces as its own PR check, so a new test/e2e-plans/*.yaml is picked up automatically without editing the workflow.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.