What does this skill do, and when should you use it?
Blog Google is a callable sub-skill in the claude-blog repository for collecting Google API data about blog performance. It covers PageSpeed Insights, CrUX, Search Console, URL Inspection, the Indexing API, GA4, Google NLP, YouTube, and Google Ads Keyword Planner. Features unlock progressively based on API-key, OAuth or service-account, GA4, and Ads credentials. The skill can support performance audits, indexation checks, search analysis, keyword research, and PDF or HTML reporting.
The skill first checks credentials in the shared file ~/.config/claude-seo/google-api.json and detects the available credential tier. Python scripts then call Google APIs for Lighthouse and CrUX checks, 25-week Core Web Vitals trends, Search Console analytics, URL inspection, Indexing API notifications, GA4 organic-traffic reports, NLP entity and sentiment analysis, YouTube search, and Google Ads keyword ideas. It can also generate chart-based PDF or HTML reports and provides handling guidance for missing credentials, unavailable CrUX data, quota exhaustion, and rate limits.
- A blog operator wants both Lighthouse lab results and real Chrome-user Core Web Vitals for a published page.
- An SEO team needs Search Console clicks, impressions, CTR, rankings, and quick-win queries ranking from positions 4 to 10.
- A site administrator needs to verify URL indexation, canonical selection, crawl state, and mobile usability.
- A content team wants relevant YouTube videos for embedding or NLP entity analysis to identify E-E-A-T gaps.
- A marketing team with GA4 or Google Ads credentials needs organic-traffic reporting or keyword-planner data.
How do you install this skill?
- Obtain explicit user confirmation before running /index or batch indexing_notify, and explain the API's official eligibility limits and irreversible external effect.
- Before installation or execution, verify that scripts/run.py, google_auth.py, and other referenced files actually exist; the supplied evidence contains path inconsistencies.
- Google API, YouTube, GA4, GSC, and Ads features may be affected by mainland-China network reachability, account permissions, quotas, and billing; “free” does not mean universally available.
- Do not commit API keys, OAuth clients, or service-account files; further restrict file permissions and API scopes.
- Shell / CLI
- Network access
- Local filesystem
Python 3.11+Google Cloud projectGoogle API key and/or OAuth/service-account credentials
Install this skill as part of the AgriciDaniel/claude-blog collection. The README provides these Claude Code marketplace commands: /plugin marketplace add AgriciDaniel/claude-blog, followed by /plugin install claude-blog@agricidaniel-claude-blog. The repository can also be cloned and installed with its supplied installer; that process applies to the collection rather than this sub-skill alone. Restart Claude Code after installation.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/AgriciDaniel/claude-blog.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/blog-google" ~/.claude/skills/
rm -rf "$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?
Start by checking credentials with python3 skills/blog-google/scripts/run.py google_auth --check --json. Supported triggers include /blog google pagespeed <url>, /blog google crux <url>, /blog google crux-history <url>, /blog google gsc <property>, /blog google inspect <url>, /blog google index <url>, /blog google ga4 [property-id], /blog google nlp <url-or-text>, /blog google youtube <query>, /blog google keywords <seed>, /blog google report <type>, and /blog google quotas. The detected credential tier should be communicated before running commands. The supplied materials do not include the full credential-setup procedure; they only direct users to references/auth-setup.md when setup is needed.
What are this skill's strengths and limitations?
- Combines page speed, real-user experience, indexation, search performance, GA4, NLP, YouTube, and keyword-planning data in one skill.
- The tiered credential model makes several useful features available with an API key alone.
- Includes 25-week CrUX history, batch URL inspection, and PDF/HTML report generation.
- Documents important operational boundaries, including Search Analytics delay, insufficient CrUX traffic, and Indexing API restrictions.
- Requires a Google Cloud project and appropriate credentials; Search Console, GA4, and keyword-planning features need additional access or account configuration.
- NLP has a 5,000-unit monthly free tier and requires GCP billing; the other APIs also impose quotas.
- The Indexing API is officially limited to JobPosting and BroadcastEvent/VideoObject pages, so it is not a general-purpose blog indexing submission tool.
- PDF generation may require system-level Pango and Cairo libraries; the supplied materials do not identify a dedicated test suite for this sub-skill.
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 |
|---|---|---|---|---|
| Blog Google: Google Data for Blog Performance this page | 47 · Use with care | ★ 2.4k | 14d ago | MIT |
| Google SEO Data Assistant | 52 · Use with care | ★ 19k | 23d ago | MIT |
| Off-Site & AI Referral Analyzer | 59 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| SEO Content Audit Skill | 48 · Use with care | ★ 2.8k | 8mo ago | MIT |
| Dark Social Attribution Analyst | 59 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
How did FollowSkills review this skill?
The evidence shows credential-tier gating, a required credential check, API-key/service-account configuration guidance, quota disclosure, and some permission guidance. However, sensitive credential paths are documented, /index and batch submission create external write effects, and per-action confirmation, revocation, rollback, and complete data-flow disclosure are incomplete, so points are deducted.
The skill documents commands, quotas, and several abnormal cases, while the supplied script handles 404, 429, request failures, and malformed values. However, most commands reference an unprovided scripts/run.py, and the authentication check conflicts with the script path in auth-setup.md. Key paths cannot be reproduced from the supplied files, so the score is substantially reduced.
Audience, commands, argument patterns, credential tiers, and major use cases are fairly clear. Evidence is missing for non-fit boundaries, protection against semantic false triggers, Chinese-language support, and mainland-China reachability; core functionality depends on Google services that may be difficult to reach from mainland China, so points are deducted.
The skill has frontmatter, MIT licensing, version metadata, argument hints, prerequisites, quick references, templates, quota documentation, and troubleshooting guidance. It lacks a clearly scoped changelog, explicit maintenance responsibility, and a clear update path for this skill, and repository/version descriptions are inconsistent, preventing a high score.
The commands have defined objectives and mostly direct JSON/report outputs, with reusable HTML/PDF templates for several SEO workflows. Static evidence does not verify representative outputs, and the command-to-script mismatch may require substantial user repair, so only limited effectiveness credit is warranted.
API endpoints, request fields, thresholds, quotas, and one trend-analysis script provide some auditability. There is no independent execution evidence for the selected skill's key paths; README test claims and CI are repository-level evidence and do not cover the referenced run.py commands. Cross-source corroboration and fact/inference separation are limited, so 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 →