# Seo Internal Links

> >

- **Type:** Skill
- **Install:** `agentstack add skill-lionkiii-claude-seo-skills-seo-internal-links`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [lionkiii](https://agentstack.voostack.com/s/lionkiii)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [lionkiii](https://github.com/lionkiii)
- **Source:** https://github.com/lionkiii/claude-seo-skills/tree/main/skills/seo-internal-links

## Install

```sh
agentstack add skill-lionkiii-claude-seo-skills-seo-internal-links
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Internal Link Audit

Crawls a domain to build an internal link graph. Identifies orphan pages, underlinked pages,
and broken internal links. Suggests anchor text improvements.

## Inputs

- `domain`: Target domain URL (e.g., `https://example.com`). Include protocol.
- `max_pages` (optional): Max pages to crawl. Default: 100. Maximum: 200.

## Execution

**Step 1: Crawl Site**

Use `scripts/fetch_page.py` and `scripts/parse_html.py` to crawl the site:

```bash
# Fetch homepage
python3 scripts/fetch_page.py 
# Parse HTML to extract links
python3 scripts/parse_html.py 
```

Follow internal links only (same domain hostname). Normalize URLs:
- Remove trailing slashes (treat `/about` and `/about/` as same)
- Remove URL fragments (`#section`)
- Preserve query strings only if they appear to be content (not tracking: strip `utm_*`, `ref=`, `source=`)

Respect robots.txt: fetch `/robots.txt` first, skip disallowed paths.
Cap crawl at `max_pages`. Track crawl queue (BFS order from homepage).

**Step 2: Build Link Graph**

For each crawled page, record:
- Source URL
- Target URL (all internal links found)
- Anchor text for each link

Build adjacency structure:
- `outbound[url]` = list of (target, anchor_text)
- `inbound[url]` = list of (source, anchor_text)

**Step 3: Calculate Per-Page Metrics**

For each crawled page:
- Inbound internal links count (links from other crawled pages pointing here)
- Outbound internal links count (links from this page to other pages)
- Link depth from homepage (BFS level at which this page was first discovered)

**Step 4: Identify Issues**

- **Orphan pages** (hard finding): Pages discovered via sitemap (`/sitemap.xml`) or linked from other pages but with **zero** inbound internal links. Fetch sitemap first to identify all known URLs.
- **Underlinked pages** (heuristic flag): Pages with fewer inbound internal links than a configurable threshold (default: 3). This threshold is a heuristic for prioritizing internal-linking work — Google defines no numeric minimum, so it is not a Google requirement.
- **Excessive outbound**: Pages with > 100 outbound internal links (PageRank dilution).
- **Broken internal links**: For top 20 most-linked pages, verify HTTP status. Flag 4xx/5xx.

**Step 5: Anchor Text Suggestions for Top 5 Underlinked Pages**

For each of the 5 most underlinked pages (fewest inbound links, excluding orphans):
1. Fetch the page and extract H1 and `` tag
2. Identify top 3 relevant anchor text options based on: H1 noun phrases, title keywords, page URL slug
3. Find pages in crawl that mention related terms and could link to this page
4. Output: target URL, suggested anchors (ranked), recommended source pages

Anchor text guidance (per Google's crawlable-links doc,
https://developers.google.com/search/docs/crawling-indexing/links-crawlable):
- Write descriptive, reasonably concise link text relevant to the target page
- Avoid generic anchors like "click here", "read more", or "website"
- Links must be `` elements with an `href` attribute resolving to a crawlable URL — Google can't reliably extract script-driven pseudo-links

**Step 6: Optional Ahrefs Enrichment**

If Ahrefs available (ToolSearch '+ahrefs'):
- Fetch `site-explorer-pages-by-internal-links` for the domain
- Cross-reference crawl findings with Ahrefs data
- Add `### Ahrefs Internal Link Data` section showing discrepancies

## Output Format

```
## Internal Link Audit: [domain]

### Summary Stats

| Metric | Value |
|--------|-------|
| Pages crawled | N / max_pages |
| Total internal links | N |
| Avg inbound links per page | N.N |
| Orphan pages found | N |
| Underlinked pages (below threshold, default 100)

| Page | Outbound Links |
|------|----------------|

## Data Sources

- Internal crawl via fetch_page.py + parse_html.py ([N] pages, [date])
- [Ahrefs MCP — if used]
```

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [lionkiii](https://github.com/lionkiii)
- **Source:** [lionkiii/claude-seo-skills](https://github.com/lionkiii/claude-seo-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-lionkiii-claude-seo-skills-seo-internal-links
- Seller: https://agentstack.voostack.com/s/lionkiii
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.