How to do the job, not just call the endpoint
Five short guides on the problems behind the APIs: verification without SMTP, on-page audits, PDF parsing at the edge, deterministic text analysis, and PII redaction. Each one says plainly where the method stops working.
Redact before the prompt, not after the leak
Where customer text escapes when it reaches a model, and how to redact with labels so the answer still makes sense.
API reference
Every route, parameter and curl example, generated from each live OpenAPI document.
- EmailCheck reference, 5 routes
- SEO Audit reference, 8 routes
- DocParse reference, 7 routes
- TextMind reference, 11 routes
- PII Shield reference, 6 routes
Verify an email address without sending a single message
What an email verification API can actually check without SMTP, how to read the score, and where the approach stops working.
The on-page audit you can run in one request
The on-page signals that decide whether a page can rank, how to read them from a single audit response, and what the audit cannot tell you.
PDF text extraction that runs where your code already runs
Why PDF parsing breaks in edge runtimes, and how to get clean text, page counts and metadata from one POST without bundling a parser.
Sentiment on every row, priced like a regex
Deterministic text analysis you can run on every row, when to pay for a model instead, and how to read the confidence numbers.
Stop sending customer text to places that keep it
Detecting emails, phones, cards and tokens in text, choosing between redact, mask and hash, and why offsets matter more than counts.