Skip to content
SkillModel-led4,210 tokens in use

seo-backlinks

Analyze a backlink profile with free or paid data sources, flag toxic links and anchor patterns, find competitor gaps and verify that known links still exist.

31,135 stars on the repository MIT LicenseAgriciDaniel/claude-seo · skills/seo-backlinks
/plugin marketplace add AgriciDaniel/claude-seo /plugin install claude-seo@agricidaniel-claude-seo
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing seo-backlinks costs

40tokens idle
4,210tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

shelldetectedwrites filesdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter11/15
  • Structure17/25
  • Examples12/15
  • Output15/15
  • Size11/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 225
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 225

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community72
  • Maturity32
  • Momentum40

seo-backlinks builds a seven-part backlink report: profile overview, anchor text distribution, referring domain quality, toxic link detection, top linked pages, competitor gap analysis and new or lost links. It detects which sources are available (DataForSEO, Moz, Bing Webmaster, Keywords Everywhere, Common Crawl and a verification crawler) and weights each by confidence.

When too little data exists it refuses to print a numeric health score and says so, and a validator script must pass before any report is shown.

What You Get

  • A backlink health score out of 100, or the words INSUFFICIENT DATA when fewer than four factors have data
  • Anchor text distribution against branded, naked, generic and exact-match benchmarks
  • Toxic link review with high-risk and medium-risk indicators and disavow candidates
  • A competitor gap list of domains that link to rivals but not to you
  • Per-link verification status such as verified, link_removed, unverifiable_js and lost

Common Use Cases

  • Review a link profile before a link-building campaign
  • Find disavow candidates after a traffic drop
  • Compare referring domains against a competitor
  • Check that links from a partner list are still live

What Makes It Unique

It will not invent a score. With only Common Crawl data it reports Not Assessed in place of every number, and a validator fails the report if a number appears without a scoreable source. Free sources work out of the box, with DataForSEO as the premium tier.

Under The Hood

The skill runs shell commands that start Python scripts through the claude-seo launcher. They call the Moz, Bing Webmaster and Keywords Everywhere APIs with keys you configure, read the Common Crawl domain graph, and fetch the pages on your link list over the network to check whether a link still exists. A validator script checks the report before it is shown. It can use DataForSEO MCP tools when present, and it offers to write a PDF report. It loads two shared reference files on demand. The frontmatter does not declare the shell or network use. Fetched content is treated as untrusted data.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers