Skip to content
Agentmart
Submit a tool

heal

Diagnose and fix failing or flaky Playwright tests in an existing suite using the executor → healer loop with root-cause classification, without weakening…

Skill by rinkucool007

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/rinkucool007-heal

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Diagnose and fix failing or flaky Playwright tests in an existing suite using the executor → healer loop with root-cause classification, without weakening assertions. Use when the user says tests are failing, broken after a UI change, flaky, timing out, or asks to "fix", "stabilize" or "self-heal" Playwright tests.

Use cases: design, testing

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/heal

What's inside

Skill name
heal
Description
Diagnose and fix failing or flaky Playwright tests in an existing suite using the executor → healer loop with root-cause classification, without weakening assertions. Use when the user says tests are failing, broken after a UI change, flaky, timing out, or asks to "fix", "stabilize" or "self-heal" Playwright tests.
License
MIT
Instructions
12 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Heal failing tests

Scope: **$ARGUMENTS** (empty = whole suite, chromium only)

1. `pw-state init "heal: $ARGUMENTS"` then `pw-state advance PLAN && pw-state advance GENERATE && pw-state advance EXECUTE` (heal-only runs skip authoring phases).
2. Delegate to **pw-executor** for the scope. If the user provided a CI `results.json`, run `pw-results <file>` instead of re-running.
3. For **flaky** reports: ask the executor to run `--repeat-each=5 --retries=0` on the suspect tests to reproduce first.
4. Loop: `pw-state advance HEAL` → `pw-state heal-attempt` → **pw-healer** → `pw-state advance EXECUTE` → **pw-executor**. Re-explore via **pw-explorer** when the healer returns `needsExplore`.
5. Stop when green, budget exhausted, or only APP_BUG / ENVIRONMENT failures remain.
6. Report: per failure → root cause, fix (file:line), verified?, and any app bugs with expected vs actual. Log each decision with `pw-state reason`.

From rinkucool007/playwright-autopilot, MIT.

More from rinkucool007/playwright-autopilot

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 30, 2026 (29 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
Repository
github.com/rinkucool007/playwright-autopilot
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Source
GitHubgithub-graphql

More from rinkucool007

Cite this page

heal skill. Agent Mart. https://agentmart.dev/skills/rinkucool007-heal (last updated 2026-10-24)