Answers to your questions.

What kinds of pages can Orsa chart?
Any public URL on the open web. Static pages, JS-rendered apps, sites behind Cloudflare, sites with infinite scroll — all of it. We handle proxy escalation, JS rendering, and bot detection so you don’t have to think about it.
How does Orsa handle sites that block bots?
We run a fleet of patched headless browsers with clean fingerprints and a tiered proxy system that escalates only when needed. We respect robots.txt by default and you can opt in to more aggressive behavior on a per-call basis.
Is there a free way to try Orsa?
Yes. You can start without a card and build against the full API with published limits in the docs.
Will my integration keep working as I grow?
Yes. We don’t deprecate access tiers. If you outgrow current limits, the API returns clear rate-limit responses so you can adjust without surprises.
Do you sell person-level data or scrape PII?
No. We are a web context layer, not a contact database. We don’t sell emails, phone numbers, or anything that identifies a person. If your use case needs that, we are not the right tool.
What happens to my old Paragon Intelligence API keys?
They keep working forever. No breaking changes, no forced migration. Update imports at your own pace.
Is there an MCP server?
Yes. Install `@orsa/mcp` or point your Claude Desktop config at our hosted MCP endpoint. Every Orsa tool becomes available inside any MCP-compatible client.
How is Orsa different from Firecrawl / Brandfetch / Clearbit?
Short version: Firecrawl is great at markdown at the low end; Brandfetch is a logo API; Clearbit was a contact database that shut down. Orsa collapses brand data, web extraction, and entity resolution into one key. Long version → the comparison pages in the nav.

More comparisons: Orsa vs alternatives.

Answers to your questions.

What kinds of pages can Orsa chart?
Any public URL on the open web. Static pages, JS-rendered apps, sites behind Cloudflare, sites with infinite scroll — all of it. We handle proxy escalation, JS rendering, and bot detection so you don’t have to think about it.
How does Orsa handle sites that block bots?
We run a fleet of patched headless browsers with clean fingerprints and a tiered proxy system that escalates only when needed. We respect robots.txt by default and you can opt in to more aggressive behavior on a per-call basis.
Is there a free way to try Orsa?
Yes. You can start without a card and build against the full API with published limits in the docs.
Will my integration keep working as I grow?
Yes. We don’t deprecate access tiers. If you outgrow current limits, the API returns clear rate-limit responses so you can adjust without surprises.
Do you sell person-level data or scrape PII?
No. We are a web context layer, not a contact database. We don’t sell emails, phone numbers, or anything that identifies a person. If your use case needs that, we are not the right tool.
What happens to my old Paragon Intelligence API keys?
They keep working forever. No breaking changes, no forced migration. Update imports at your own pace.
Is there an MCP server?
Yes. Install `@orsa/mcp` or point your Claude Desktop config at our hosted MCP endpoint. Every Orsa tool becomes available inside any MCP-compatible client.
How is Orsa different from Firecrawl / Brandfetch / Clearbit?
Short version: Firecrawl is great at markdown at the low end; Brandfetch is a logo API; Clearbit was a contact database that shut down. Orsa collapses brand data, web extraction, and entity resolution into one key. Long version → the comparison pages in the nav.

More comparisons: Orsa vs alternatives.