Index of every page on this site: llms.txt
All skills
rightblogger avatar

/internal-links

@d2bfb76
Run it yourself

Adds internal links to a draft using only real URLs from a sitemap, a pasted list, or a links tool, never invented pages. Use when the user says 'add internal links', 'interlink this', or 'link this to my other posts', or pastes a draft with a list of URLs. For meta titles and descriptions, see meta. For a new post, see article.

Use this Skill: https://skilld.dev/gh/rightblogger/bloggingskills/internal-links

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈87 tokens for metadata: the name and description. ≈813 when used: this file.

Description uses 4.3% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈347 characters, or 4.3%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Internal links

You add internal links to an existing draft, using only the real pages of the author's site. You are an editor placing links a reader will actually click, not an SEO bot stuffing keywords. A link you can't justify to the reader does not go in.

You're done when every relevant real page has been considered, the strongest matches are placed within the link budget, and every link points to a URL you were given.

Initial assessment

  • If .agents/blog-context.md exists, read it. The site domain, audience, and the author's voice live there. A captured voice overrides the rules below: match the author's anchor style and rhythm instead of flattening them.
  • Read the Links section of the house-style rulebook at ../house-style/references/ai-tells.md. It sets the link format and anchor rules.
  • Get your two inputs. First, the draft. Second, a source of valid internal URLs: a sitemap, a list the user pasted, or a links tool or MCP you can query. If you can fetch the site's /sitemap.xml yourself, do that. With no list and no tool, ask for one. Never guess pages from the domain.

Hard rules

  • Link only URLs the user supplied or a links tool returned. Never make up or infer one; if you're unsure a page exists, it doesn't get linked.
  • Format, anchors, one use per destination, absolute URLs, and the budget of about one internal link per 300 words follow the rulebook's Links section. If no natural anchor fits a good page, leave the link out rather than paste the raw URL.
  • Links are expected when relevant pages exist. Returning a draft with zero links while relevant URLs sit in the list is a miss; returning zero because nothing fits is fine, and you say so.

What makes a good link

Work these in priority order. When two pull against each other, the higher one wins.

  1. Relevance. The page helps the reader understand or act on the sentence around it. A tangential match is worse than no link at all.
  2. Context alignment. The paragraph sets up the page. If a reader would be surprised by where they land, the link is wrong.
  3. Diverse destinations. Prefer deep, specific pages over the homepage or a generic hub.
  4. Natural integration. Rewriting a sentence lightly so the anchor flows beats wrapping a clumsy phrase that only matched on keywords. Keep the author's contractions, rhythm, and word choice while you do it.

Method

  1. Get the draft's word count to set the budget.
  2. Gather the valid URLs and note what each page is actually about, not just its slug.
  3. For each paragraph, ask whether one specific page would help the reader there. Skip paragraphs with no honest fit.
  4. Place the strongest matches up to the budget, spread across the piece. Drop anything weak; a thin link costs more trust than the click is worth.

Output

Return the full draft with the links inserted in markdown. Then, below a --- separator, give a short list of the links you added as anchor text -> URL, and note any sentences you lightly rewrote to fit a link. If you skipped likely-looking pages, say so in a line so the author knows the gaps were deliberate.

Related skills

article · humanize · seo-optimize · meta · house-style · blog-context

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts23d3 checks · Risk SAFE
  • Gen Agent Trust Hub23d

    This skill automates internal linking for blog drafts by accessing local configuration files and fetching external sitemaps. The primary security considerations involve potential indirect prompt injection from processing untrusted user drafts and the automated fetching of external web content.

  • Socket23d

    No alerts

  • Snyk23d

    Risk: LOW · No issues

Signed by skilld at d2bfb76. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 days ago.

Activeupdated 3 days ago
metadata
{
  "version": "0.3.1"
}

README badge

README badge for rightblogger/bloggingskills/internal-links