full-seo-audit
Skill3 starsChanged 5 months agoMITView on GitHub
Run a complete technical-SEO audit of a website by orchestrating every individual audit (sitemap, robots.txt, redirects, internal links, external links, metadata, headings, images, schema markup, canonical tags, Open Graph, mixed content, pagination, soft 404s, content quality, keyword cannibalization, site architecture, Core Web Vitals, and llms.txt) and consolidating the results into one prioritized report. Use when the user asks for a full, complete, comprehensive, end-to-end, or site-wide SEO audit, a technical SEO review, or a single report covering many SEO issues at once.
- Installs packages
In this repository19 other files it gives its agents.
- canonical-tag-auditSkill
- content-quality-auditSkill
- core-web-vitals-auditSkill
- external-link-auditSkill
- heading-structure-auditSkill
- image-seo-auditSkill
- internal-link-auditSkill
- keyword-cannibalization-auditSkill
- llms-txt-auditSkill
- meta-data-auditSkill
- mixed-content-auditSkill
- open-graph-auditSkill
- pagination-auditSkill
- redirect-auditSkill
- robots-txt-auditSkill
- schema-markup-auditSkill
- site-architecture-auditSkill
- sitemap-auditSkill
- soft-404-auditSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Run the individual SEO audit skills, then merge their findings into one consolidated, prioritized report.
- Distinctive: line 14
Use this when the user wants the whole picture rather than one area:
- Distinctive: line 28
Focus: run everything (default), or a chosen subset (e.g., "just crawlability and indexation").
- Very specific: line 110
Write the report following references/report-template.md:
- Very specific: line 138
If the user wants the report as a .docx (for example, to share with stakeholders or attach to a ticket), save the Markdown report to a file …
- Very specific: line 148
If the user wants the raw findings as a spreadsheet (for filtering, sorting, or triage in Sheets or Excel), convert the audit's …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Full SEO Audit
Run the individual SEO audit skills, then merge their findings into one consolidated, prioritized report. This skill is the orchestrator; the real work is done by the per-area skills it coordinates.
When to use this skill
Use this when the user wants the whole picture rather than one area:
- "Run a full / complete / comprehensive SEO audit of example.com"
- "Do a technical SEO review of my site"
- "Give me one report covering everything that's wrong with my SEO"
For a single area ("check my sitemap", "audit my titles"), use that area's skill directly instead.
Inputs to collect
- Site URL: the live site root (e.g.,
https://example.com). - Scope: how many pages to crawl (default ~200) and how many sitemap URLs to probe. Larger sites should be sampled.
- Focus: run everything (default), or a chosen subset (e.g., "just crawlability and indexation"). Map the user's intent to the skills below.
- API key (optional): Core Web Vitals field data needs a Google PageSpeed Insights key; without it, that audit uses lab proxies only.
Coverage map
Each row is a separate skill. Run the ones in scope.
| Area | Skill | Crawls / fetches |
| Indexation: sitemaps | sitemap-audit | XML sitemaps |
| Crawlability: robots.txt | robots-txt-audit | robots.txt |
| Crawlability: architecture | site-architecture-audit | site + robots + sitemap |
| Redirects | redirect-audit | URL list / crawl |
| Internal links | internal-link-audit | crawl |
| External links | external-link-audit | crawl |
| Canonical tags | canonical-tag-audit | crawl |
| Pagination | pagination-audit | crawl |
| Soft 404s | soft-404-audit | crawl |
| Titles & descriptions | meta-data-audit | crawl |
| Headings | heading-structure-audit | crawl |
| Images | image-seo-audit | crawl |
| Structured data | schema-markup-audit | crawl |
| Social metadata | open-graph-audit | crawl |
| Content quality | content-quality-audit | crawl |
| Keyword cannibalization | keyword-cannibalization-audit | crawl |
| HTTPS / mixed content | mixed-content-audit | crawl |
| Page experience | core-web-vitals-audit | per-URL |
| AI discoverability | llms-txt-audit | /llms.txt |
Workflow
Step 1: Crawl once into a shared page cache
The page-based audits all read the same HTML, so crawl the site a single time with the shared fetch stage and let every extractor read from that cache:
mkdir -p audits
python3 scripts/fetch_pages.py https://example.com --max-pages 200 --output audits/page_cache.jsonStep 2: Extract + audit each in-scope area
For each page-based area in scope, run its extractor with --from-cache (pure,
no extra crawl) then its audit script. Save each audit output to the shared
folder using a stable, skill-named filename so they can be merged:
# repeat per in-scope page-based skill: all read the ONE page_cache.json
python3 ../meta-data-audit/scripts/extract_metadata.py --from-cache audits/page_cache.json --output audits/meta.inventory.json
python3 ../meta-data-audit/scripts/audit_metadata.py audits/meta.inventory.json --output audits/meta-data-audit.audit_report.json
python3 ../heading-structure-audit/scripts/extract_headings.py --from-cache audits/page_cache.json --output audits/head.inventory.json
python3 ../heading-structure-audit/scripts/audit_headings.py audits/head.inventory.json --output audits/heading-structure-audit.audit_report.jsonThe skills that read --from-cache are: meta-data, heading-structure,
image-seo, schema-markup, open-graph, content-quality, keyword-cannibalization,
mixed-content, canonical-tag, pagination, soft-404, and external-link. (A few
also run a small secondary probe on top of the cached crawl, for canonical
targets, share images and the soft-404 missing-page probe.)
The non-page audits fetch their own thing and don't use the page cache: run them
per their own SKILL.md: sitemap-audit, robots-txt-audit, llms-txt-audit,
redirect-audit, core-web-vitals-audit, plus internal-link-audit and
site-architecture-audit (which crawl with depth tracking of their own).
Step 3: Consolidate
Merge every *.audit_report.json into one cross-skill view:
python3 scripts/consolidate_reports.py --dir audits --output consolidated_report.jsonThe consolidator emits total issues by severity, a per-skill breakdown, and a flat list of every failing check ranked by severity then count.
Step 4: Produce the consolidated report
Write the report following references/report-template.md:
- Executive summary: overall health, issue counts by severity, the 3–5 highest-impact fixes across all areas.
- Scorecard: one row per audited area with its issue counts and a status.
- Findings by area: a short section per skill summarizing its top issues and linking to (or inlining) that skill's own detailed report.
- Prioritized roadmap: a single cross-area action list ordered by impact, not by area.
Step 5: Prioritize across areas
Rank fixes by impact, not by which skill found them. Apply this order:
- Indexation blockers first: anything stopping pages from being crawled or
indexed (robots
Disallow: /, sitewide noindex, sitemap full of dead URLs, broken canonicals) outranks on-page polish. - Sitewide/template issues over one-off issues: a bug in one template affecting 500 pages beats a single page's long title.
- High-traffic, high-intent pages first within a given fix.
- Group related findings across skills (e.g., the same
http→httpsmigration surfaces in sitemap, redirect, internal-link, canonical, and mixed-content audits: present it once as a single root-cause fix).
Optional: export the report
As a Word document (.docx)
If the user wants the report as a .docx (for example, to share with stakeholders or attach to a ticket), save the Markdown report to a file and convert it:
python3 scripts/md_to_docx.py report.md --output report.docxscripts/md_to_docx.py uses only the Python standard library (no pip install) and renders headings, tables, lists, links, bold/italic, and code blocks. Offer this whenever a user asks for a Word doc, a .docx, or a shareable/downloadable report.
As a CSV of findings (.csv)
If the user wants the raw findings as a spreadsheet (for filtering, sorting, or triage in Sheets or Excel), convert the audit's audit_report.json directly:
python3 scripts/findings_to_csv.py audit_report.json --output findings.csvscripts/findings_to_csv.py is also standard-library only. It writes one row per finding, with check and severity columns prepended and list fields (e.g. the pages sharing a duplicate value) joined with ; . Unlike the .docx, which reformats the written report, the CSV is a direct dump of the structured findings: offer it whenever a user wants the data itself, a spreadsheet, or to slice findings by check or severity.
Resources
scripts/md_to_docx.py: convert the Markdown report into a Word (.docx) document (standard library only)scripts/findings_to_csv.py: flatten the audit findings JSON into a CSV, one row per finding (standard library only)references/report-template.md: consolidated report structurereferences/coverage-map.md: what each sub-skill covers and what it needsscripts/consolidate_reports.py: merge per-skill audit reports into one ranked report
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

