Auditing & CI/CD Tools

Tools built on Plasmate for site auditing, accessibility testing, and CI/CD integration.

Site Auditor

Crawl and audit websites for SEO and content quality issues. 10x faster than Chrome-based tools like Screaming Frog.

pip install plasmate plasmate-audit

# Audit a site (crawls up to 50 pages)
plasmate-audit https://example.com

# More pages, JSON output
plasmate-audit https://example.com --max-pages 200 --json

Python API:

from plasmate_audit import audit_site

result = audit_site("https://example.com", max_pages=50)
print(f"Score: {result['score']}/100")
print(f"Errors: {result['errors']}, Warnings: {result['warnings']}")

Checks: missing titles, heading structure, empty links, thin content, images without alt text.

GitHub

Accessibility Auditor

Analyze what a screen reader would "see" using Plasmate's Semantic Object Model.

pip install plasmate plasmate-a11y

python -m plasmate_a11y https://example.com

Python API:

from plasmate_a11y import audit_url

result = audit_url("https://example.com")
for issue in result['issues']:
    print(f"[{issue['severity']}] {issue['message']}")

Checks: heading hierarchy, image alt text, link text quality, form labels, page language, landmark regions.

GitHub

GitHub Action

Fetch web pages with Plasmate in your CI/CD workflows.

- uses: plasmate-labs/som-action@v1
  with:
    url: https://example.com
  id: som

- run: echo "Page title: ${{ steps.som.outputs.title }}"

Outputs: som (full SOM JSON), title (page title), tokens (token count).

Use cases:

Verify deployed content after deploy
Monitor page structure changes
Content regression testing
Automated accessibility checks in CI

GitHub

Jupyter Notebooks

Interactive notebooks for learning Plasmate. Run locally or on Google Colab.

pip install plasmate jupyter tiktoken pandas matplotlib
git clone https://github.com/plasmate-labs/notebooks
cd notebooks
jupyter notebook

Notebooks:

Getting Started - explore SOM structure
Token Comparison - measure HTML vs SOM across 10 sites
Batch Extraction - process 20+ URLs in parallel
Research Agent - build a web research pipeline

GitHub

Replit Template

Try Plasmate in your browser with zero setup.

GitHub