agentleFS
strikersamstrikersam/autonomous-ai-agency/.claude/skills/seo-audit-report/SKILL.md

seo-audit-report

Skill12 starsChanged 29 days agoMITView on GitHub

Full-site SEO / GEO / AEO / AIO audit for any website. Crawls up to N pages using Chrome TLS impersonation (bypasses Akamai/Cloudflare), runs a Screaming Frog-class check catalog across Technical, Content, Security, Social, GEO, and AIO pillars, and produces an executive-level PDF report plus CSV/JSON/Markdown data files. Suitable for client deliverables and CTO reviews.

  • Installs packages

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is 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.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run a full SEO / GEO / AEO / AIO audit against any public website and produce an executive-level PDF report suitable for CTO or client …

    line 49
  2. Very specific:

    All other parameters have sensible defaults.

    line 65
  3. Distinctive:

    Before trusting results, spot-check the JSON:

    line 92
  4. Distinctive:

    Verify bypass success: check the JSON as shown above.

    line 129

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill: seo-audit-report

Purpose

Run a full SEO / GEO / AEO / AIO audit against any public website and produce an executive-level PDF report suitable for CTO or client delivery.

The crawl uses curl_cffi with Chrome 120 TLS impersonation to bypass bot-protection layers (Akamai, Cloudflare) that block plain httpx crawlers. Playwright browser mode is supported when system dependencies are available; curl_cffi is the reliable fallback.

Quick Start

python scripts/run_seo_audit.py \
  --website-url https://www.example.com \
  --output-dir ./seo-output

All other parameters have sensible defaults. See --help for full options.

Parameters

ParameterDefaultDescription
--website-urlrequiredSite to audit
--max-pages100Crawl budget (pages)
--max-depth3Link-follow depth
--output-dir./seo-audit-outputWhere to write files
--monthly-organic-revenue0Revenue baseline for at-risk modelling
--check-image-sizestrueHEAD-request image sizes (slower)
--no-check-image-sizes—Skip image size checks for speed

Output Files

FileContents
{domain}_seo_audit.pdfExecutive PDF report (cover, executive summary, pillar deep-dives, full findings, delegation plan, methodology)
{domain}_report.jsonFull machine-readable report (all pages, all issues, pillar scores)
{domain}_report.mdMarkdown version of the report
{domain}_findings.csvSummary CSV: one row per finding type with priority, URLs affected, how-to-fix
{domain}_pages.csvPer-page CSV: URL, status, title, H1, word count, issues found
{domain}_issues.csvPer-issue-instance CSV: URL × check_code × details

Verify Bypass Quality

Before trusting results, spot-check the JSON:

import json
with open("seo-output/example.com_report.json") as f:
    r = json.load(f)
pages = r["pages"]
# All should show status_code=200 and via="curl_cffi"
print({p["url"]: (p["status_code"], p.get("fetch_via","?")) for p in pages[:5]})

Repeated 403s with small HTML bodies indicate the site is still blocking; report that plainly rather than fabricating findings.

How This Skill Works (Agent Instructions)

  1. Install dependencies (first run only):
    cd autonomous-ai-agency
    python -m venv .venv && source .venv/bin/activate
    pip install -r requirements.txt
    # playwright install chromium  # optional — curl_cffi is used as fallback
  2. Run the audit:
    source .venv/bin/activate
    python scripts/run_seo_audit.py \
      --website-url "{{ website_url }}" \
      --max-pages {{ max_pages | default(100) }} \
      --max-depth {{ max_depth | default(3) }} \
      --output-dir "{{ output_dir | default('./seo-audit-output') }}" \
      --monthly-organic-revenue {{ monthly_organic_revenue | default(0) }}
    Expect 3–20 minutes depending on site size and network. For large sites (500 pages) expect up to 90 minutes with browser mode.
  3. Verify bypass success: check the JSON as shown above.
  4. Present the PDF to the user and provide a 3–5 sentence summary:
    • Overall health score and worst pillar
    • Top 3 issues by strategic pressure
    • Pages crawled vs pages genuinely bypassed vs still blocked

Revenue-at-Risk Disclaimer (load-bearing — always include in reports)

Revenue-at-risk figures use a saturating exponential model share = 0.35 × (1 − e^(−pressure/50)) applied to aggregate issue severity. They are model estimates of proportional organic revenue exposure, not measured losses. Leave monthly_organic_revenue=0 when the baseline is unknown; all figures will display as £0/month.

Troubleshooting

SymptomCauseFix
All pages return 403Bot wall not bypassedcurl_cffi should handle most; try increasing timeout_seconds
libXdamage.so.1 errorMissing Playwright system libscurl_cffi fallback activates automatically
Audit times outSite too large / slowReduce --max-pages or --max-depth
Report has 0 pagesCrawl seeding failedCheck URL is correct and publicly reachable

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.