searchcans-reader-seo-auditlisted

# SearchCans Reader SEO Audit Extract a public URL and report what the Reader API returned plus observable page signals. Do not equate a successful extraction with indexability, ranking, accessibility compliance, or legal permission to reuse content. ## Run a minimal audit first Set `SEARCHCANS_API_KEY` in the execution environment. Never expose it in a report or commit. Request HTML when canonical, H1, description, or JSON-LD checks are needed: ```bash python scripts/reader_page_audit.py "https://example.com/article" \ --include-html --out page-audit.json ``` Use `--headless --wait-ms 3000` only for a page whose important content is rendered by JavaScript. For a PDF or Office-document URL, add `--file`. Use `--screenshot 1` or `--screenshot 2` when a visual artifact is needed. ## Escalate carefully Start with `--proxy 0`. If the result is empty or blocked, retry with the next proxy tier and record the smallest tier that works. Do not automatically escalate every URL to a higher-cost tier. The default `--account-mode auto` skips a pre-flight check for standard Reader extraction and enforces one before a higher-cost proxy request. Use `--account-mode warn` to record the account state without blocking, `enforce` to block insufficient work, `cap` (equivalent to enforce for one URL), or `off` to skip the account check. The result includes a sanitized `account_guard` summary only; never expose the raw Account API response. Read `references/audit-interpretation.md` befo