The Model Context Protocol (MCP) provides an open framework that allows LLMs to interact with external tools and services through well-defined server endpoints. WebVeta leverages MCP to expose its search and retrieval infrastructure directly to AI agents, enabling developers and researchers to build intelligent, context-aware applications without writing custom API wrappers or managing backend infrastructure.
WebVeta allows 4 tools through MCP Servers.
- webveta_text_search – Provides text search results i.e and links to search results
- webveta_neural_search – provides LLM based responses. ($200, $1000 tiers)
- webveta_page_context – provides Page Insights for a given URL ($1000 tier)
- webveta_page_plain_text – provides clean text of a webpage, a curl and parsing of HTML can be done by LLMs but cleaned text more useful.
*** webveta_neural_search *** could increase LLM costs, but there are cost control mechanisms of max cost per day and caching of responses. The rest are counted as part of search request limits but does not incur additional costs. If necessary don’t allow webveta_neural_search.
The following JSON snippet can be used in software such as Claude-Desktop, LM Studio, Cherry Studio etc… allows using WebVeta for www.alightservices.com and sub-domains.
{
"mcpServers": {
"webveta-clientapi": {
"url": "https://1.api.webveta.alightservices.com/",
"headers": {
"X-WebVeta-DataId": "d-e03e3b2e-df07-11f0-8fcb-000d3a9bba43"
}
}
}Here is an example system prompt for efficiently using, but there could be various efficient system prompts.
* You are a an expert MUST always use the search tool before answering any question.
* For every user query, you MUST perform at least two (2) distinct initial searches using the search tool, with related but different query formulations.
* After the initial searches, analyze all search results carefully.
* You MUST then perform at least one (1) follow-up search that refines or deepens the query based on the earlier search results.
* Use the search tool additional times as needed until you have collected enough information to answer confidently.
* Always base your answers ONLY on information from the search tool responses, not on your own assumptions or prior knowledge.
* If the search tool requires a dataid or dataId parameter, always pass an empty string ("") for that parameter.
* Do not skip searches, even if you think you already know the answer; you must still perform the required searches first.
–
Mr. Kanti Kalyan Arumilli
B.Tech, M.B.A
Founder & CEO, Lead Full-Stack .Net developer
ALight Technology And Services Limited
Phone / SMS / WhatsApp on the following 3 numbers:
+91-789-362-6688, +1-480-347-6849, +44-07718-273-964
kantikalyan@gmail.com, kantikalyan@outlook.com, admin@alightservices.com, kantikalyan.arumilli@alightservices.com, KArumilli2020@student.hult.edu, KantiKArumilli@outlook.com and 3 more rarely used email addresses – hardly once or twice a year.
