™2 min read
Crawl policy
The portal’s robots.txt explicitly allows GPTBot, ClaudeBot, PerplexityBot and Google-Extended, and it ships an llms.txt summary and an ai.txt usage policy. Crawlers are welcome — the goal is accurate coverage, not exclusion.
Structured data
Every content page carries a JSON-LD graph — Organization, WebSite, BreadcrumbList and Article or FAQPage — so a crawler can parse the meaning, not just the text. Breadcrumbs are threaded to the apex so the estate reads as one site.
Live grounding via MCP
Beyond static content, the MCP server lets an agent query products, quotes and eligibility live. That is the strongest form of GEO: the model reads the current answer from the source rather than a cached page.
Frequently asked questions
Do you block AI crawlers?
No — we welcome them. The robots.txt allows GPTBot, ClaudeBot, PerplexityBot and Google-Extended, and we ship llms.txt and ai.txt so AI answers about Credicorp are accurate.
What is the best way for an AI to get current Credicorp facts?
The MCP server. It lets an agent read products, quotes, eligibility and tiers live from the source, rather than relying on a crawled snapshot.
Funding for UK limited companies
Credicorp lends to your company, not to you personally — short-term working capital with no personal guarantee. See what your business could access.