e2e-pr-stabilizer

# E2E PR Stabilizer Stabilize the Playwright E2E suite for a single pull request using **evidence, not assumptions**. Spans, traces, and the live app are the source of truth — not the CI dashboard. This skill never proposes a fix without a measurement to point at, and never commits a fix until three consecutive local runs prove it works. > **This `SKILL.md` is a thin index.** > Detailed procedures live in [`rules/*.md`](./rules) and [`templates/*.md`](./templates). > Each phase loads only what it needs. --- ## What this skill combines | Source | Role | |--------|------| | [Playwright Healer agent](https://playwright.dev/docs/test-agents) — external; one of the [Playwright Test Agents](https://playwright.dev/docs/test-agents) on the [Playwright MCP server](https://github.com/microsoft/playwright-mcp) | Test-debugging methodology — how to fix a Playwright test correctly. | | [`/playwright-trace-analyzer`](../../analysis/playwright-trace-analyzer/SKILL.md) | Per-run `trace.zip` extraction, hotspot ranking. | | [`/ci-auto-fix`](../../delivery/ci-auto-fix/SKILL.md) | Reused only for Phase 7's single push + watch — the iteration loop no longer lives here. | | Dash0 MCP server (`dash0-dev` or `dash0-prod`) | Historical evidence — failure recurrence, retry counts, span-level evidence across CI runs. | | Local Playwright runner | Primary evidence source — trace.zip per run, OTel spans to Dash0 (`ci.is_ci=false`), and the live app for selector verification. | | GitHub Actions (one...