agentsclimarketplace

Linking and schema

Skill siddiqss/semantic-seo-suite/skills/linking-and-schema

Grounded semantic SEO, GEO & off-page as Claude Code skills — with a fabrication guard that refuses to invent numbers. Free & MIT.

Install
npx -y skills add siddiqss/semantic-seo-suite --skill linking-and-schema

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 6 stars6 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Generate an internal-linking plan and JSON-LD structured data from a brand's topical map and entity profile. Use whenever the user asks about internal links, anchor text, site structure, linking strategy, schema markup, structured data, or JSON-LD — or right after new content is drafted or published. Produces an actionable link checklist (with varied anchors) and validated per-node schema files. Triggers on linking/schema intent broadly.

SKILL.md

3.2 KB, as published. Nobody here has run it

linking-and-schema

Turn the map into a link graph and structured data. Both outputs are deterministic and carry the framework's fact discipline: laterals are justified (not sprayed), anchors are varied (not exact-match spam), and schema never states an ungrounded fact.

Read first: ../../framework/internal-linking-rules.md, ../../framework/schema-jsonld.md, ../../framework/eeat-signals.md.

Preconditions

  • topical-map.json and entity-profile.json for the brand.
  • Optional: map embeddings (for lateral justification) and crawl data (for real orphans).

Workflow

  1. Internal-linking plan:

    python ../../scripts/linking_plan.py --map brands/<slug>/topical-map.json \
  [--vecs brands/<slug>/data/embeddings/map.npz --lateral-threshold 0.72] \
  [--crawl-dir brands/<slug>/data/crawl] --brand "<Brand>" \
  --out brands/<slug>/audits/linking-plan.md

    For each node it emits up/down/justified-lateral links with varied, descriptive anchors drawn from the target's query network, plus link-health flags (planned-graph orphans, over-linked hubs, and real crawl orphans if crawl data is present). Laterals below the embedding threshold are marked REVIEW rather than dropped or trusted.

  2. JSON-LD schema:

    python ../../scripts/schema_jsonld.py --map brands/<slug>/topical-map.json \
  --entity-profile brands/<slug>/entity-profile.json \
  --locked brands/<slug>/locked-facts.json \
  --out-dir brands/<slug>/data/schema

    Emits one .jsonld per node (HowTo / FAQPage / Product / Article), a site-level Organization graph, and a checklist.md. Product Offer prices come ONLY from locked facts; FAQ/HowTo answer text is left as a placeholder to be filled from the validated draft — so no invented answer ships in markup.

  3. Hand off: give the user the link checklist (page → insert link → suggested anchor → where) and the schema files. Note that syntactic validation here is necessary but not sufficient — run Google's Rich Results Test before publishing.

Definition of done

  • Link plan covers every node; laterals justified; orphans/over-links flagged.
  • One schema file per node + org graph; 0 validation issues; no fabricated price/rating.
  • Anchors are varied across placements (no exact-match repetition).

Grounding ladder

  • T0: link plan from map structure; laterals justified by named shared attribute (asserted). Schema from map + profile.
  • T1: laterals justified by embedding distance (derived); real orphans from crawl.
  • Author Person schema is emitted only when a real author/credential exists (eeat-signals.md) — never fabricated to fill the template.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.