skills/content/wordpress-live-validation/SKILL.md
Validate published WordPress posts in browser via Playwright.
npx skillsauth add notque/claude-code-toolkit wordpress-live-validationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill loads a real WordPress post in a Playwright headless browser and verifies that what readers see matches what was uploaded. The browser is the source of truth -- REST API success does not guarantee correct rendering.
Browser backend selection: Playwright MCP (default) is used for automated validation and CI/CD. Use Chrome DevTools MCP when the user explicitly asks to "check in my browser" or "debug live", or when the task involves Lighthouse audits or performance profiling.
| Signal | Load These Files | Why |
|---|---|---|
| tasks related to this reference | validation-checks.md | Loads detailed guidance from validation-checks.md. |
| tasks related to this reference | playwright-tools.md | Loads detailed guidance from playwright-tools.md. |
| tasks related to this reference | phase-checks.md | Loads detailed guidance from phase-checks.md. |
| tasks related to this reference | output-format.md | Loads detailed guidance from output-format.md. |
| example-driven tasks | examples.md | Loads detailed guidance from examples.md. |
| errors, error handling | error-handling.md | Loads detailed guidance from error-handling.md. |
Read-Only Observation Only: Never click, type, fill forms, or modify anything on the WordPress site. This is observation-only validation—any write action risks mutating published content.
Evidence-Based Reporting: Every check result must reference a concrete artifact (DOM value, network response, screenshot path). "Looks fine" is not acceptable. Report what the browser shows, not assumptions.
Non-Blocking Reports: Failed validation produces a report but does not revert the upload or block the pipeline. The user decides how to act on findings.
Severity Classification (enforce strictly):
Browser Availability: Requires either Playwright MCP or Chrome DevTools MCP. If neither is available, exit in Phase 1 with a skip report. Do not retry.
Default Behaviors (ON):
article → .entry-content → .post-content → mainOptional Behaviors (OFF unless enabled):
Input Requirements:
{WORDPRESS_SITE}/?p={post_id}&preview=true)Goal: Load the WordPress post and confirm the content area is present.
See references/phase-checks.md "Phase 1: NAVIGATE — Detailed Steps" for the 4-step procedure (verify browser MCP availability, navigate, wait for content area with selector chain, remove cookie/consent banners with the JS snippet).
GATE: Page loaded with HTTP 200 (or 30x redirect to 200), content selector found. If 4xx/5xx or no selector found: capture screenshot, report FAIL with HTTP status, STOP. Do not proceed to Phase 2.
Verification means execution, not reasoning. Run the command. Do not reason about whether the command would pass. Do not summarize the expected output. Execute the check, paste the exit code, paste the relevant output. A verification phase that produces a verdict without an observed tool result is not a verification — it is a guess with a rigor aesthetic.
Goal: Inspect rendered DOM and network activity for content integrity and SEO completeness. Run all checks without stopping on individual failures.
The 7 checks (with severity, JS extraction snippets, and pass/fail criteria) are documented in references/phase-checks.md "Phase 2: VALIDATE — All 7 Checks":
GATE: All 7 checks executed, each with severity and evidence. Proceed to Phase 3.
Goal: Verify rendering at three standard breakpoints. Capture visual evidence.
Test each viewport in sequence:
| Viewport | Width | Height | Represents | |----------|-------|--------|------------| | Mobile | 375 | 812 | iPhone-class | | Tablet | 768 | 1024 | iPad-class | | Desktop | 1440 | 900 | Standard laptop |
See references/phase-checks.md "Phase 3: RESPONSIVE CHECK — Detailed Steps" for the per-viewport 4-step procedure (resize, screenshot, overflow check, container visibility check) with the JS snippets.
GATE: Screenshots captured at all three viewports. Overflow and visibility recorded. Proceed to Phase 4.
Goal: Produce structured pass/fail report with severity counts and evidence artifacts.
See references/output-format.md for the full report template, status markers ([PASS], [FAIL], [WARN], [INFO], [SKIP]), and result classification rules.
GATE: Report generated with accurate severity counts. Screenshots saved. Result matches blocker tally.
When invoked after wordpress-uploader, this skill acts as an optional Phase 5: POST-PUBLISH VALIDATION. See references/examples.md for the integration flow and the post_url / post_id / --title inputs the uploader provides.
The validation is non-blocking by default: a FAIL result produces a report for the user but does not revert the upload. The user decides whether to act on findings.
See references/examples.md for worked examples (post-upload validation, standalone live check, OG tag verification) and the wordpress-uploader integration flow.
See references/error-handling.md for the full error matrix (Playwright MCP not available, page returns 4xx/5xx, content selector not found, network timeout on image checks, cookie banner blocks content).
For detailed check specifications and Playwright tool usage:
documentation
Document translation: quick/normal/refined modes with chunked parallel subagents and glossary support.
development
AI image generation: Gemini and Nano Banana backends; single/series/batch workflows with prompt-to-disk.
testing
Unified voice content generation pipeline with mandatory validation and joy-check. 13-phase pipeline: LOAD, GROUND, STATS-CHECKPOINT, GENERATE, HOOK-GATE, VALIDATE, REFINE, VARIETY-GATE, JOY-CHECK, ANTI-AI, CLOSE-GATE, OUTPUT, CLEANUP. Use when writing articles, blog posts, or any content that uses a voice profile. Use for "write article", "blog post", "write in voice", "generate content", "draft article", "write about".
documentation
Critique-and-rewrite loop for voice fidelity validation.