What does this skill do, and when should you use it?
This is the DevOps skill bundled in the Playwright repository, focused on CI failure analysis, workflow debugging, and release operations. It provides a guide for analyzing GitHub Actions failures associated with the latest commit on main, plus a shell script for downloading failed job logs. The script stores logs under ~/tmp/commit-<sha>/. It does not represent the browser testing or automation capabilities of the repository’s sibling skills.
Analyzes GitHub Actions failures for the last commit on main; runs fetch-commit-logs.sh to download failed job logs; writes downloaded logs to ~/tmp/commit-<sha>/; provides the CI Commit Failure Report guide.
- A Playwright maintainer needs to investigate why the latest commit on main failed in GitHub Actions.
- An engineer debugging CI workflows needs to retrieve logs from failed jobs for local analysis.
- A release-operations contributor needs the repository’s documented DevOps guidance when investigating workflow problems.
How do you install this skill?
- Before running, verify that gh is authorized with only the repository and Actions read access required.
- Failure logs may contain tokens, environment variables, or other sensitive data; assess redaction, access control, and retention before storing them.
- The guide and script disagree about handling in-progress jobs, and failure feedback for missing gh, jq, network access, or authentication is unspecified.
- The script writes under $HOME/tmp, so environment differences and path permissions may affect portability.
- Shell / CLI
- Network access
- Local filesystem
Use the repository’s .claude/skills/playwright-devops/ directory as the skill directory. The supplied material does not document a separate installation command for this skill; the README only states that the broader skill collection can optionally be installed with playwright-cli install --skills.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/microsoft/playwright.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/.claude/skills/playwright-devops" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Analyze the GitHub Actions failures for the last commit on main.
Give a compatible client a request such as: "Analyze the GitHub Actions failures for the last commit on main." When logs are needed, run fetch-commit-logs.sh; it downloads failed job logs into ~/tmp/commit-<sha>/. The source does not specify a more detailed trigger format or a complete release procedure.
What are this skill's strengths and limitations?
- Clearly scoped to CI failure analysis and workflow debugging.
- Includes a log-fetching script that collects failed job logs in a predictable directory.
- Uses standard YAML frontmatter with `name` and `description`.
- Ships as part of the Playwright repository, which is licensed under Apache-2.0.
- The supplied skill content contains only one guide and one script, so the detailed diagnostic workflow is limited.
- The log-download workflow requires shell access, network access, and a local filesystem.
- The source does not provide an independent installation procedure, test coverage, or platform validation details.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Playwright DevOps this page ✓ Microsoft · Official | 44 · Not recommended | ★ 98k | 12d ago | Apache-2.0 |
| GitHub PR Check Fixer ✓ OpenAI · Official | 47 · Use with care | ★ 28k | 4mo ago | — |
| CI Push Approval Assistant ✓ OpenAI · Official | 45 · Not recommended | ★ 129k | 12d ago | Apache-2.0 |
| SmokedMeat E2E Testing Skill | 49 · Use with care | ★ 393 | 1mo ago | AGPL-3.0 |
| React Native GitHub Actions Artifacts | 49 · Use with care | ★ 1.7k | 18d ago | MIT |
How did FollowSkills review this skill?
The skill explicitly queries GitHub APIs and writes failed logs locally; the behavior appears read-oriented with no malware, covert exfiltration, or destructive default visible. However, it relies on the user's existing gh authorization, gives no least-privilege guidance or confirmation, does not address sensitive data in logs, and provides no cleanup or rollback mechanism, so points are deducted.
SKILL.md, the guide, and the shell script define a coherent fetch-analyze-report flow. Reliability is reduced because bash, gh, jq, network access, and authentication are assumed without availability checks, diagnostic failure handling, or tests. The guide says in-progress workflows and jobs are included, but the script filters jobs to conclusion == failure, creating an observable inconsistency.
The name, description, optional SHA input, and report goal clearly target GitHub Actions CI analysis for the Playwright repository. Boundaries for non-fit cases, malformed input, permission or network failures, semantic trigger precision, Chinese-language use, and mainland-China reachability are not specified, so points are deducted.
The skill has a compact structure with a main file, guide, and script, while repository context supplies Apache-2.0 licensing, version information, and Microsoft provenance. The skill itself lacks dependency/install notes, a precise parameter contract, troubleshooting examples, limitation disclosure, changelog/versioning, and an explicit maintenance update path, so points are deducted.
The documented workflow can produce a directly usable Markdown report grouped by workflow, job, and test failure. Effectiveness is reduced because the implementation does not match the stated inclusion of in-progress jobs, root-cause classification is not verified, and no representative output or execution evidence is present in the reviewed files.
The script, report template, and repository metadata provide auditable primary source material and limited traceability. There are no skill-specific tests, CI coverage, or third-party execution records demonstrating the key path, so independent verification remains thin and points are deducted.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →