razbakov/daily-seo-review

Daily SEO Review

Automated daily SEO health check for two production sites:

| Site | Domain | Project Path | |------|--------|-------------| | dancegods | dancegodscompany.com | ~/Projects/dancegodscompany/engineering/website | | razbakov.com | razbakov.com | ~/Projects/ikigai/engineering/razbakov.com |

Trigger

• Scheduled daily at 5am via ~/.claude/scheduled-tasks/daily-seo-review/SKILL.md
• Manual: "run daily SEO review" or /daily-seo-review

Process

Run for each site in parallel where possible.

1. Check Sitemap

# Try sitemap_index first, then sitemap.xml
curl -s "https://[domain]/sitemap_index.xml" | head -30
curl -s "https://[domain]/sitemap.xml" | head -30

# Count total URLs
curl -s "https://[domain]/sitemap.xml" | grep -c "<url>" || \
  curl -s "https://[domain]/sitemap_index.xml" | grep -c "<sitemap>"

Flag if: sitemap missing, returns 4xx/5xx, or URL count drops vs yesterday.

2. Check Robots.txt

curl -s "https://[domain]/robots.txt"

Check:

• Sitemap reference present
• No critical pages disallowed (/, /blog, /about, /style)
• AI crawlers: GPTBot, ClaudeBot, Google-Extended — note if blocked

3. Check Meta Tags on Key Pages

for url in "/" "/about" "/blog"; do
  echo "=== $url ==="
  curl -sL "https://[domain]${url}" | \
    grep -oE '<title>[^<]+</title>|<meta[^>]*(name="description"|property="og:title"|property="og:image")[^>]*>' \
    | head -10
  echo
done

Check:

• Every page has unique <title>
• description meta present (150–160 chars)
• og:title and og:image present

4. Check Canonical and Hreflang

for url in "/" "/about" "/blog"; do
  echo "=== $url ==="
  curl -sL "https://[domain]${url}" | \
    grep -oE '<link[^>]*(canonical|hreflang)[^>]*>' | head -20
  echo
done

Check:

• <link rel="canonical"> present on each page
• Canonical URL matches served URL (no trailing slash mismatch)
• hreflang alternates cover all supported locales
• x-default hreflang present

5. Check Structured Data (JSON-LD)

curl -sL "https://[domain]/" | python3 -c "
import sys, re
html = sys.stdin.read()
schemas = re.findall(r'<script[^>]*type=[\"\\']application/ld\+json[\"\\'][^>]*>(.*?)</script>', html, re.DOTALL)
print(f'Found {len(schemas)} JSON-LD schemas')
for s in schemas[:3]:
    print(s[:300])
"

Expected schemas:

• Homepage: WebSite, WebPage
• Blog post: BlogPosting, BreadcrumbList
• About: Person, AboutPage

6. Check HTTP Status + Prerendering

for url in "/" "/about" "/blog"; do
  status=$(curl -sI "https://[domain]${url}" | head -1 | awk '{print $2}')
  cache=$(curl -sI "https://[domain]${url}" | grep -i cache-control | head -1)
  echo "$status  $url  —  $cache"
done

Flag: any non-200/301 status, missing cache-control: public (may indicate SSR).

7. Check Indexing Proxy (Google Search Console)

Since GSC API requires authentication, do a quick spot-check:

# Check if pages are discoverable (site: operator equivalent check)
curl -sL "https://[domain]/sitemap.xml" | grep -c "<loc>"

Document for human review: "Visit GSC → Coverage → check Excluded / Not Indexed count." Primary concern: dancegods has 97% pages not indexed — track via GSC manually until fixed.

Report Format

Save to [project_path]/reports/seo-daily-YYYY-MM-DD.md:

# Daily SEO Report — [Domain] — YYYY-MM-DD

## Summary
| Check | Status | Notes |
|-------|--------|-------|
| Sitemap | OK/FAIL | [N] URLs found |
| Robots.txt | OK/WARN/FAIL | |
| Meta tags | OK/WARN/FAIL | |
| Canonical/hreflang | OK/WARN/FAIL | |
| Structured data | OK/WARN/FAIL | [N] schemas |
| HTTP status | OK/FAIL | |

**Overall: HEALTHY / NEEDS ATTENTION / CRITICAL**

## Issues Found
- [ ] [CRITICAL] [description] — [suggested fix]
- [ ] [WARNING] [description] — [suggested action]

## Indexing Status (manual GSC check required)
- dancegods: ~97% not indexed (known issue — track canonical/trailing slash fix)
- razbakov.com: [status]

## Raw Check Output
[paste curl outputs]

Reporting After Both Sites

1. Append combined summary to today's daily review session file: ikigai/sessions/YYYY-MM-DD-daily-review.md
2. If CRITICAL issues found: create a GitHub issue in the relevant repo and reference the report
3. No Telegram send on routine OK — only on CRITICAL or new regressions

Integration Points

• nuxt-seo-audit — deeper one-off audit (this skill is the daily lightweight check)
• content-seo-agent — fixes and content work (triggered when issues found)
• daily-review — morning review incorporates SEO status
• ikigai/sessions/ — reports linked from daily review files
• dancegods reports/ — ~/Projects/dancegodscompany/engineering/website/reports/
• razbakov.com reports/ — ~/Projects/ikigai/engineering/razbakov.com/reports/