API + MCPWeb Search API + MCP Server · For Developers Building AI Agents

Web Search API & MCP for AI Agents

Real-time web search and page content for AI agents — every result ships with verifiable evidence, not just links.

Start for free
searchpipe — live pipeline
➜ POST /search {"query":"How to deploy FastAPI","include_answer":true}
SEARCH ──▶ RETRIEVE ──▶ RERANK ──▶ RETURN 2.41s
# 200 OK — 2.41s · 2 credits
"answer": "FastAPI is typically served with uvicorn, often behind gunicorn…[0][2][3]",
"answer_audit": { "score": 0.92 },
"results": [
{ "url": "https://fastapi.tiangolo.com/deployment/", "score": 0.98,
"evidence": { "quote": "…deployed behind uvicorn or gunicorn…", "source_tier": "official" } },
{ "url": "https://uvicorn.org/deployment/", "score": 0.94 }
] // 5 results · verbatim evidence · auto-refund on failure
Definition

What is SearchPipe?

SearchPipe is a web search API and MCP server for developers building AI agents. It provides web search and web content retrieval through a single interface, so AI applications can access information from the live web.

Every result ships with verifiable evidence — span-level quotes, claim verdicts, and source authority — not just links.

Primary category: Web Search API
Also: Web Search Infrastructure, MCP Server


Pipeline

How SearchPipe Works

Your agent sends a single query; we run search, extraction, scoring and summarization in the background, and return clean results ranked by relevance.

01

Search

Search the web for the requested information across multiple engines.

02

Retrieve

Retrieve relevant web pages and extract clean article content.

03

Rerank

Rank results based on relevance using LLM scoring.

04

Return

Return structured results to the application or AI agent.

05

Verify

Every result ships with span-level quotes, source authority and fetch metadata, and audited answers add per-statement claim verdicts.


Capabilities

Web retrieval infrastructure for AI agents

01

Multi-engine retrieval

Aggregated results from multiple search engines on self-hosted infrastructure.

02

Full-text extraction

Fetches top-N pages and extracts the article body, stripping navigation and ads.

03

LLM relevance reranking

Every result scored 0–1 against the query intent and sorted by relevance.

04

AI answers with citations

Optional summary with citation markers, flagged ai_generated=true.

05

Evidence spans

Every result carries a verbatim supporting quote with character offsets from the fetched page text.

06

Claim verdicts

Audited answers break into per-statement verdicts — supported, partial, unsupported or inconclusive — with an overall audit score.

07

URL health

Each evidence item reports fetch status code, redirects and the final URL of the cited page.

08

Source authority

Sources are classified into six tiers: official, press, academic, preprint, community, unknown.

09

Remote MCP server

streamable-http MCP endpoint; tool name searchpipe_search, key in the URL.

10

REST API

POST /search callable from any language, framework or backend service.

11

Automatic refunds

Failed calls (502 upstream failure, 400 output violation) are refunded idempotently.

12

Rate limiting

Redis sliding window: 100 requests per minute with a burst of 20 by default.

13

Result caching

Identical query and parameters served from cache for 300 seconds by default.

GET STARTED

Start with 1,000 free credits every month

No card required. Get your API key and make your first call in under 3 minutes.

Get API Key →

Interfaces

Two first-class ways to connect

MCP and the REST API are equal first-class paths to the same pipeline — same parameters, same results, same billing. Pick whichever fits your stack; switching later costs nothing.

Web Search API

POST /search · any language

Use the REST API when integrating SearchPipe directly into your application or agent infrastructure. One POST /search endpoint, callable from any language, framework or backend: authenticate with Authorization: Bearer sp-YOUR_API_KEY and get structured, scored JSON your code consumes directly.

curl -X POST https://searchpipe.tech/v1/search \
  -H "Authorization: Bearer sp-YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query":"FastAPI tutorial","max_results":5,"output_type":"answer"}'
import httpx

resp = httpx.post(
    "https://searchpipe.tech/v1/search",
    headers={"Authorization": "Bearer sp-YOUR_KEY"},
    json={"query": "FastAPI tutorial", "max_results": 5, "output_type": "answer"},
    timeout=60,
)
resp.raise_for_status()
data = resp.json()
print(data["answer"])
for r in data["results"]:
    print(r["score"], r["title"], r["url"])
const resp = await fetch("https://searchpipe.tech/v1/search", {
  method: "POST",
  headers: {
    "Authorization": "Bearer sp-YOUR_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ query: "FastAPI tutorial", max_results: 5, output_type: "answer" }),
});
if (!resp.ok) throw new Error(`Search failed (${resp.status}): ${await resp.text()}`);
const data = await resp.json();
console.log(data.answer);
for (const r of data.results) console.log(r.score, r.title, r.url);
package main

import (
	"bytes"
	"encoding/json"
	"fmt"
	"io"
	"net/http"
	"time"
)

func main() {
	payload, _ := json.Marshal(map[string]any{
		"query":          "FastAPI tutorial",
		"max_results":    5,
		"output_type": "answer",
	})
	req, _ := http.NewRequest("POST", "https://searchpipe.tech/v1/search", bytes.NewReader(payload))
	req.Header.Set("Authorization", "Bearer sp-YOUR_KEY")
	req.Header.Set("Content-Type", "application/json")
	client := &http.Client{Timeout: 60 * time.Second}
	resp, err := client.Do(req)
	if err != nil {
		panic(err)
	}
	defer resp.Body.Close()
	body, _ := io.ReadAll(resp.Body)
	fmt.Println(string(body))
}
import java.net.URI;
import java.net.http.*;
import java.time.Duration;

public class SearchPipeDemo {
    public static void main(String[] args) throws Exception {
        String json = """
            {"query":"FastAPI tutorial","max_results":5,"output_type":"answer"}
            """;
        HttpClient client = HttpClient.newBuilder()
            .version(HttpClient.Version.HTTP_1_1)
            .connectTimeout(Duration.ofSeconds(10))
            .build();
        HttpRequest request = HttpRequest.newBuilder()
            .uri(URI.create("https://searchpipe.tech/v1/search"))
            .timeout(Duration.ofSeconds(60))
            .header("Authorization", "Bearer sp-YOUR_KEY")
            .header("Content-Type", "application/json")
            .POST(HttpRequest.BodyPublishers.ofString(json))
            .build();
        HttpResponse<String> resp =
            client.send(request, HttpResponse.BodyHandlers.ofString());
        System.out.println(resp.body());
    }
}
using System.Net.Http.Json;

using var client = new HttpClient();
client.DefaultRequestHeaders.Authorization =
    new System.Net.Http.Headers.AuthenticationHeaderValue("Bearer", "sp-YOUR_KEY");

var payload = new
{
    query = "FastAPI tutorial",
    max_results = 5,
    output_type = "answer",
};

HttpResponseMessage resp = await client.PostAsJsonAsync("https://searchpipe.tech/v1/search", payload);
resp.EnsureSuccessStatusCode();
Console.WriteLine(await resp.Content.ReadAsStringAsync());

Self-serve for agents

Send the link to your AI agent and it will complete the whole MCP setup by itself: /agent-setup/SKILL.md.

Production-grade guardrails

Input/output content moderation, per-call billing with automatic refunds on failure, and rate limiting — safe even for high-frequency agent calls.

Identical behavior on both paths

Cache, moderation, refunds and result shape are shared — an agent connected via MCP and a service calling REST get byte-identical results for the same query.


Use Cases

Built for developers building AI agents

SearchPipe is designed for developers who need to give AI agents and AI applications access to web search and web content retrieval.

Research Agents

Search the web for current information and gather relevant sources. Agents can autonomously explore topics, synthesize findings, and cite live URLs without hallucinating outdated facts.

Coding Agents

Give coding agents access to documentation, repositories and current technical information. Let Claude Code, Cursor or your custom dev agent look up APIs, libraries and solutions on the fly.

RAG Applications

Retrieve fresh web content to augment LLM context. Replace stale vector stores with real-time extraction and reranking, ensuring your generation is always grounded in the latest information.

AI Applications

Add web search capabilities without building your own search infrastructure. Integrate via REST API or MCP in minutes, with automatic refunds, rate limiting and structured JSON responses.

Deep Research Agents

Long-horizon research runs live or die on source quality. Every intermediate result carries verbatim evidence and source authority, so the agent can weigh sources instead of trusting links — and audited answers report per-statement support.

Compliance & Audit

Regulated teams cannot ship hallucinated citations. Verify each claim against its cited URL before publishing, keep the per-statement audit trail with quotes and fetch metadata, and flag unsupported statements automatically.


Response

Web Search API Example

This request searches the web and returns structured results that can be consumed by an AI application.

01

One call, ready to use

No more gluing together search APIs, crawlers, extractors and rerankers — hand the whole data pipeline to us.

02

First-class REST API

Building your own agent framework or backend service? A single POST /search endpoint runs the same pipeline with identical parameters — no MCP required.

Parameter reference, response fields, error codes and multi-language examples: API documentation.

{
  "query": "How to deploy FastAPI",
  "answer": "FastAPI is typically served with uvicorn,
    often behind gunicorn…[0][2][3]",
  "answer_audit": {
    "score": 0.92,
    "statements": [
      { "text": "…", "verdict": "supported", "confidence": 0.97 }
    ]
  },
  "results": [
    {
      "url": "https://…",
      "title": "…",
      "content": "Summary snippet…",
      "score": 0.98,
      "evidence": {
        "quote": "…verbatim supporting span…",
        "span_start": 412, "span_end": 488,
        "source_tier": "official"
      }
    }
  ]
}

MCP

MCP Example

Connect an MCP-compatible AI client to SearchPipe to give the agent access to web search.

# Add SearchPipe to any MCP-compatible agent
claude mcp add --transport http searchpipe \
  "https://searchpipe.tech/mcp/?api_key=sp-YOUR_API_KEY"

# The agent can now call searchpipe_search
# with query, max_results, include_answer and search_depth parameters

Compare

SearchPipe and Other Web Search APIs

SearchPipe is one approach to providing web retrieval for AI agents. Developers may also evaluate other web search APIs depending on their retrieval requirements.

Scope

What SearchPipe is and is not

SearchPipe is web search infrastructure for AI agents and applications.

SearchPipe is not a consumer search engine. It provides search and retrieval capabilities that developers can integrate into their own AI applications and agents.


Pricing

Simple, transparent credit pricing

1 credit = 1 request. 1,000 free credits every month — get your agent running before you pay.

$0.005 / credit

Free tier

For integration testing
$0
  • 1,000 free credits every month
  • Full API & MCP capabilities
  • Great for integration testing
Sign up now

Monthly plans

Subscription credits are valid for 30 days from grant. Subscriptions are managed by the platform and renew automatically; upgrades are supported anytime.

Starter

Monthly plan
$10/mo
  • 2,500 credits / month
  • Unused credits expire at period end
Subscribe

Max

Monthly plan
$35/mo
  • 10,000 credits / month
  • Unused credits expire at period end
Subscribe

Subscription rules: renewed credits take effect from the next billing period (they don't stack into the current one); upgrading charges the full current price of the new plan — new-plan credits land immediately with a fresh 30-day window, and unused credits from the old plan are extended to the new period end; downgrades are not supported. Credit validity and refund policies follow the Terms of Service. Full pricing and rules: Pricing. Comparing options? See how SearchPipe compares with Tavily, Exa and Brave.


FAQ

FAQ

Frequent questions on integration, billing and result usage; the full list lives on the FAQ page.

What is SearchPipe?

SearchPipe is a web search API and MCP server for developers building AI agents. A single call runs multi-engine retrieval, full-text fetching, LLM reranking and optional summarization, returning structured results with relevance scores that agents can consume directly.

How is SearchPipe different from calling a search engine directly?

SearchPipe returns cleaned page content, relevance scores and optional answers — not raw search-result HTML that you have to parse. Agents can consume the response directly, and failed requests are automatically refunded.

How do I add SearchPipe to my AI agent?

Two first-class paths, equally supported — pick whichever fits your stack: connect via MCP (any MCP-compatible agent or client — Claude Code, Cursor and more), or call the REST API POST /search directly from your custom agent in any language. Both run the same pipeline with the same search pipeline and credit billing. MCP excludes REST answer streaming and REST request metadata.

How does pricing work?

Credits: $0.005 = 1 credit for custom recharges (bulk tiers from $0.0040). Fast search costs 0.5 credits, basic 1 credit, and advanced 2 credits. The free plan includes 1,000 credits per UTC calendar month, renewed when a verified account first signs in that month. Recharged credits never expire; subscription credits are valid for 30 days.

How long does integration take?

Copy the MCP link from the dashboard into your client config — typically 1–3 minutes. Agents can also read /agent-setup/SKILL.md to configure themselves automatically.

Does SearchPipe provide a web search API?

Yes. SearchPipe provides a REST API (POST /search) that returns structured JSON results with relevance scores, extracted page content and optional AI-generated answers. Authenticate with an sp- prefixed API key.

Does SearchPipe support MCP?

Yes. SearchPipe runs a remote MCP server (streamable-http) at /mcp/. Any MCP-compatible agent or client can connect by adding the URL with an API key. Tool name: searchpipe_search.

How is SearchPipe different from a consumer search engine?

SearchPipe is not a consumer search engine. It provides search and retrieval capabilities that developers can integrate into their own AI applications and agents.

What does SearchPipe return?

Structured JSON with query, results (each containing url, title, content, score and optional raw_content), an optional answer with citation markers, and an ai_generated flag. The response shape is identical whether you call the REST API or the MCP server.

Put the live web inside your agent, today

1,000 free credits every month — connect via MCP or call the REST API in minutes. Same pipeline, same billing, your choice.