SEO Backlinks
by agricidaniel / claude-seo
Analyze a site's backlink profile, anchors, toxic signals, competitors, gaps, and disavow candidates.
Expert guide for off-page SEO and backlink analysis.
Listed ·Updated
Install data from skills.sh
SEO Off Page Backlinks is a Link Building skill for AI agents, published in the seoskills.sh catalog. Reach for it when your work involves off-page signals, digital PR, and backlink outreach that holds up. Install it with one command and it runs inside your own agent, so the work happens in your workflow, not a separate SEO tool.
Not all backlinks are equal. Evaluate every backlink across these six dimensions before determining its value.
| Dimension | Weight | What to Evaluate |
|---|---|---|
| Relevance | 25% | Is the linking site/page topically related to your content? |
| Authority | 20% | What is the linking domain's authority (DR/DA)? |
| Placement | 15% | Where on the page does the link appear? (editorial body > sidebar > footer) |
| Anchor Text | 15% | Is the anchor text descriptive, branded, or spammy? |
| Follow Status | 10% | Is the link dofollow, nofollow, sponsored, or UGC? |
| Traffic | 15% | Does the linking page receive organic traffic? |
Relevance is the most important factor. A link from a relevant DR 40 site often outperforms a link from an irrelevant DR 80 site.
Tier 1 (Best): Same niche, same topic, same audience
Tier 2 (Good): Related niche, overlapping audience
Tier 3 (OK): General authority site (news, education, government)
Tier 4 (Weak): Unrelated niche, no audience overlap
Tier 5 (Risk): Completely irrelevant, different language, spammy vertical
Highest: In-content editorial link (surrounded by relevant text)
High: Author bio on a relevant guest post
Medium: Resource page / curated list
Low: Sidebar widget or blogroll
Lowest: Footer link, site-wide link, comment link
MCP Tool: Use analyze_backlinks with a target URL to retrieve the full backlink profile with per-link quality metrics.
| Metric | Provider | Scale | Based On |
|---|---|---|---|
| Domain Rating (DR) | Ahrefs | 0-100 | Quantity and quality of backlinks to the domain |
| Domain Authority (DA) | Moz | 0-100 | Link profile, MozRank, MozTrust, and other factors |
| Authority Score | Semrush | 0-100 | Backlinks, organic traffic, and spam signals |
| Score Range | Interpretation | Typical Sites |
|---|---|---|
| 80-100 | Elite authority | Google, Wikipedia, NYT, government domains |
| 60-79 | Very strong | Major brands, popular publications, large media |
| 40-59 | Strong | Established businesses, popular niche blogs |
| 20-39 | Moderate | Small businesses, newer sites with some traction |
| 10-19 | Low | New sites, small blogs, thin content sites |
| 0-9 | Very low | Brand new domains, parked domains, spam sites |
MCP Tool: Use analyze_domain_authority to retrieve DR/DA scores along with trust signals and spam indicators for any domain.
Anchor text distribution is a critical ranking signal and a common penalty trigger when over-optimized.
| Type | Definition | Example (for a page about "best running shoes") |
|---|---|---|
| Branded | Company/site name | "Nike", "RunnerWorld.com" |
| Exact-match | Exact target keyword | "best running shoes" |
| Partial-match | Contains target keyword + other words | "guide to the best running shoes for beginners" |
| Generic | Non-descriptive text | "click here", "read more", "this article" |
| Naked URL | Raw URL as anchor | "https://runnerworld.com/best-shoes" |
| Branded + keyword | Brand combined with keyword | "RunnerWorld's best running shoes list" |
| Image link | Image used as link (alt text = anchor) | <img alt="running shoes comparison chart"> |
Target these approximate ratios for a natural-looking profile:
| Anchor Type | Target % | Risk if Over-Indexed |
|---|---|---|
| Branded | 30-40% | Low risk |
| Naked URL | 15-20% | Low risk |
| Generic | 10-15% | Low risk |
| Partial-match | 10-15% | Medium risk if excessive |
| Branded + keyword | 5-10% | Medium risk |
| Exact-match | 3-8% | High risk if >10% |
| Image/other | 5-10% | Low risk |
| Strategy | Difficulty | Scalability | Link Quality | Time to Results |
|---|---|---|---|---|
| Content marketing (linkable assets) | Medium | High | High | 2-6 months |
| Digital PR / data studies | High | Medium | Very High | 1-3 months |
| Broken link building | Medium | Medium | Medium-High | 1-2 months |
| Resource page outreach | Low-Medium | Medium | Medium-High | 2-4 weeks |
| HARO / journalist queries | Medium | Low | Very High | 1-4 weeks |
| Guest posting | Low-Medium | Medium | Medium | 2-4 weeks |
| Unlinked brand mentions | Low | Low | High | 1-2 weeks |
Create content specifically designed to attract links:
Combine data-driven content with journalist outreach:
Find pages that mention your brand but do not link to you:
"your brand" -site:yourdomain.com in GoogleSee LINK_BUILDING_PLAYBOOKS.md for detailed step-by-step playbooks with outreach templates.
Toxic backlinks can suppress your rankings or trigger manual actions. Identify them early.
| Signal | Severity |
|---|---|
| Link from a known PBN (private blog network) | High |
| Link from a site in a completely unrelated language | Medium-High |
| Link from a link farm (thousands of outbound links, no real content) | High |
| Exact-match anchor from low-quality directory | Medium |
| Sitewide footer/sidebar link from unrelated site | Medium |
| Link from a hacked or malware-flagged domain | High |
| Link from a site with DR 0-5 and no traffic | Low-Medium |
| Link from adult/gambling/pharma site (if unrelated to your niche) | High |
Paid link with no rel="sponsored" attribute | High |
| Comment spam links | Low |
See TOXIC_LINK_PATTERNS.md for 20+ detailed toxic patterns with detection criteria and recommended actions.
Your site: example.com
Competitor A: competitor-a.com → 450 referring domains
Competitor B: competitor-b.com → 380 referring domains
Competitor C: competitor-c.com → 520 referring domains
Gap domains (link to 2+ competitors but not you): 85 domains
Filtered (DR 30+, relevant): 42 domains
Priority targets: 15-20 domains
| Factor | Weight |
|---|---|
| Links to 3+ competitors (not just one) | High priority |
| DR 40+ with real organic traffic | High priority |
| Topically relevant to your niche | Required |
| Resource page or "best of" list format | Easier outreach |
| Recently published or updated content | More likely to respond |
MCP Tool: Use analyze_backlinks on competitor domains to export their backlink profiles for gap analysis.
Disavow links only when:
Do NOT disavow when:
The disavow file is a plain text file (.txt) uploaded to Google Search Console.
# Disavow file for example.com
# Generated: 2026-02-26
# Reason: Manual action - unnatural inbound links
# Individual URLs to disavow
https://spamsite.com/page-with-link.html
https://another-spam.com/directory/listing.html
# Entire domains to disavow (use domain: prefix)
domain:linkfarm-network.com
domain:pbn-cluster-site.net
domain:cheap-directory-spam.org
.txt disavow file| Pattern | Interpretation |
|---|---|
| Gradual, steady increase over months | Natural — content gaining traction organically |
| Sudden spike followed by sustained plateau | Typical for viral content or major PR hit |
| Sudden spike followed by complete stop | Suspicious — possibly purchased links |
| Perfectly linear growth (same # of links per week) | Unnatural — automated or purchased |
| Burst of links from same C-class IP range | PBN or coordinated scheme |
| Seasonal spikes aligned with industry events | Natural — correlates with real-world activity |
While internal linking is technically "on-page," its role in distributing link equity (PageRank) across your site is a critical off-page consideration.
External backlinks → Homepage (high authority)
→ Key landing pages
→ Pillar content
Homepage → Pillar pages (via navigation + in-content links)
Pillar pages → Cluster/supporting pages (via contextual links)
Cluster pages → Related cluster pages + back to pillar (hub & spoke)
| Tool | Use For |
|---|---|
analyze_backlinks | Full backlink profile analysis, competitor gap analysis, link quality scoring |
analyze_domain_authority | DR/DA scores, trust signals, spam indicators |
See LINK_QUALITY_SCORING.md for the detailed 6-dimension link scoring model. See LINK_BUILDING_PLAYBOOKS.md for step-by-step playbooks with outreach templates. See TOXIC_LINK_PATTERNS.md for 20+ toxic backlink patterns with detection criteria.
Not using the CLI? Copy the SKILL.md and paste it straight into ChatGPT, Claude, or any agent.