fix(cli): rewrite CLI command in installed skills

[Skn0tt]

When skills are installed through npx/yarn/pnpm playwright cli (or init-skills), the examples still said playwright-cli, which isn't on PATH after 1.62. Sources keep the bare name so GitHub search still finds them; install rewrites the command to the invoker's package-manager form and leaves skill name, allowed-tools, and .playwright-cli/ alone.

Fixes #42135

[github-actions]

Test results for "MCP"

1 failed
❌ [firefox] › mcp/cli-devtools.spec.ts:173 › response-body returns just the body @mcp-windows-latest-firefox

8062 passed, 1284 skipped

---

Merge workflow run.

[github-actions]

🟢 The one failure is a pre-existing Firefox flake, unrelated to this PR

Hi, I'm the Playwright bot and I took a look at the CI failure.

The single failure — [firefox] › mcp/cli-devtools.spec.ts:173 › response-body returns just the body on mcp-windows-latest-firefox — is a pre-existing flake, not something this PR caused. This PR only rewrites the CLI command embedded in installed skills; it doesn't touch the devtools/response-body path the test exercises. The PR looks clear.

Details

Overall: one failed test in the merged MCP report, Firefox-Windows only. Nothing in the diff reaches it.

Pre-existing flake / infra

• [firefox] › mcp/cli-devtools.spec.ts:173 › response-body returns just the body (@mcp-windows-latest-firefox) — flake. Across the test-results DB this test fails only on the mcp-windows-latest-firefox bot, and it does so on SHAs this PR can't be responsible for:

  • a push to main (no PR), sha 1e6a5e6 — run 30530696560
  • PR feat(chromium): roll to r1235 #41974, sha 30b37d7 (timed out) — run 30061238899
  • PR fix(mcp): launch the Chrome profile that has the extension installed #41939, sha 23ae27b — run 29954678992

  On that bot the test is 395 passed / 4 failed (~1%), and it passes on every other project (chromium, chrome, msedge, webkit: 0 failures). The PR's changes are confined to CLI skill installation (installSkills.ts, cli-client/cli-daemon program.ts, SKILL.md), which the response-body devtools flow doesn't touch. No retries in this run, so it showed red rather than being rescued — but the history says flake.

_{Triaged by the Playwright bot - agent run}

[github-actions]

Test results for "tests 1"

8 flaky

⚠️ [chromium-library] › library/popup.spec.ts:260 › should not throw when click closes popup `@chromium-ubuntu-22.04-arm-node20`
⚠️ [chromium-library] › library/video.spec.ts:495 › screencast › should capture static page in persistent context @smoke `@chromium-ubuntu-22.04-arm-node20`
⚠️ [chromium-library] › library/video.spec.ts:664 › screencast › should capture full viewport `@chromium-ubuntu-22.04-arm-node20`
⚠️ [chromium-library] › library/trace-viewer.spec.ts:1469 › should update highlight when typing snapshot `@realtime-time-library-chromium-linux`
⚠️ [chromium-library] › library/video.spec.ts:664 › screencast › should capture full viewport `@chromium-ubuntu-22.04-node20`
⚠️ [chromium-library] › library/chromium/chromium.spec.ts:299 › should report intercepted service worker requests in HAR `@chromium-ubuntu-22.04-node22`
⚠️ [firefox-page] › page/page-emulate-media.spec.ts:144 › should keep reduced motion and color emulation after reload `@firefox-ubuntu-22.04-node20`
⚠️ [playwright-test] › ui-mode-trace.spec.ts:827 › should update state on subsequent run `@windows-latest-node22`

51229 passed, 1190 skipped

---

Merge workflow run.