Skip to content
SkillModel-led3,535 tokens in use

geo-schema

Audits a page's Schema.org structured data for AI discoverability, scores it out of 100 and generates ready-to-paste JSON-LD.

11,792 stars on the repository MIT Licensezubair-trabzada/geo-seo-claude · skills/geo-schema
curl -fsSL https://raw.githubusercontent.com/zubair-trabzada/geo-seo-claude/main/install.sh | bash
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing geo-schema costs

26tokens idle
3,535tokens 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

shelldeclared and detectednetworkdeclaredwrites filesdeclared

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
  • Frontmatter8/15
  • Structure22/25
  • 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−2

    does this on the machine, as the frontmatter declares

    SKILL.md, line 31

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity100
  • Maintenance52
  • Community68
  • Maturity32
  • Momentum40

geo-schema fetches a page, finds every JSON-LD, Microdata and RDFa block, validates it against Schema.org and checks what is missing for the site’s business type. The focus is entity recognition: how clearly AI search systems can tell what an organization or person is and which other profiles belong to it.

It scores the result on a 0 to 100 rubric, audits the sameAs links across platforms such as Wikipedia, Wikidata, LinkedIn and GitHub, flags deprecated schemas, and writes GEO-SCHEMA-REPORT.md with ready-to-paste JSON-LD for each gap.

What You Get

  • A GEO-SCHEMA-REPORT.md with a schema score out of 100 and a table of detected schemas
  • Per-schema validation results covering JSON syntax, types, required and recommended properties and nesting
  • A sameAs audit across Wikipedia, Wikidata, LinkedIn, YouTube, X, GitHub, Crunchbase and other platforms
  • Ready-to-paste JSON-LD in the @graph pattern for Organization, WebSite, BreadcrumbList and business-specific types
  • A list of deprecated or changed schemas with replacements

Common Use Cases

  • Checking whether a company site has a complete Organization entity with sameAs links
  • Adding Article and author markup to a blog to strengthen authorship signals
  • Generating Product or SoftwareApplication markup for a store or SaaS site
  • Finding JSON-LD that only appears after JavaScript runs and so may be processed late

What Makes It Unique

The skill frames schema work around AI discoverability, where most schema guides target Google rich results. It gives specific attention to sameAs strategy with a ranked platform list, to knowsAbout and speakable properties, and to whether markup sits in server-rendered HTML. It also warns that a generic web fetch strips the head and loses JSON-LD, and sends the agent to a page-fetch script that returns parsed structured data.

Under The Hood

The skill is written instructions for the agent with no scripts in its own folder. It tells the agent to run the fetch_page.py script from the main geo skill through the shell to download the page and read its parsed structured_data, because a standard web fetch drops the head section. The agent may also fetch other pages and check that URLs inside the markup resolve, so it reaches out over the network to the site you name and the links in its schema. It then writes GEO-SCHEMA-REPORT.md into the working directory. It reads the page and reports; it does not change the site itself, and the generated JSON-LD is left for you to paste in.

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