Audit any URL from a terminal.
nmav wraps the public REST API so you can run AI-readiness audits from a terminal or a CI/CD pipeline. Same engine as the web app, same daily quota, one binary, zero runtime dependencies.
# install globally npm install -g @nymanmedia/ai-visibility-cli # or run ad-hoc npx @nymanmedia/ai-visibility-cli audit https://example.com
Requires Node.js 20.11 or newer.
# .github/workflows/ai-readiness.yml
- name: AI visibility audit
env:
NMAV_API_KEY: ${{ secrets.NMAV_API_KEY }}
run: npx -y @nymanmedia/ai-visibility-cli audit "$PREVIEW_URL" --fail-on-score 70 --json > audit.json--fail-on-score N exits non-zero when the audit score is below N, so the job fails the build on a regression. --json writes a machine-readable report you can ship to an artifact store.
Search engines and AI assistants treat your site differently than they did five years ago. Crawlers parse JSON-LD, follow llms.txt, respect or ignore robots.txt rules differently per agent. A regression in any of those signals is invisible until your traffic moves. The CLI lets you make AI-readiness a build-gate. Caught at PR review, not at quarter-end.
0· success3· rate-limited (HTTP 429)1· generic error4·--fail-on-scorethreshold not met2· usage error (missing arg, bad flag)