Crawlora
Open navigation
Sign inTry Playground Console

Developer guides

OpenAI Agents Integration Guide

Use Crawlora endpoints as callable tools so agents can retrieve structured public web data from supported platforms.

Tool callingStructured outputsServer-side keysResearch agents
Browse APIs Try PlaygroundView AI Agent Guide

Verified HTTP pattern

POST /google/search

Normalized JSON

Request

POST https://api.crawlora.net/api/v1/google/search
x-api-key: $CRAWLORA_API_KEY
Content-Type: application/json

{
  "country": "us",
  "keyword": "best CRM software",
  "language": "en",
  "limit": 10,
  "page": 1
}

Base URL

https://api.crawlora.net/api/v1

Auth header

x-api-key

Example endpoint

POST /google/search

Exact agent SDK imports change over time, so this page uses framework-neutral tool-calling patterns and labels schemas as illustrative.

Developer workflow

Why agents need structured web data

Agents can browse or call APIs. For repeatable workflows, structured API outputs are easier to validate, summarize, and store than arbitrary page HTML.

Developer workflow

Good Crawlora tools for agents

  • google_search.
  • google_maps_search.
  • youtube_transcript.
  • youtube_comments.
  • tiktok_search.
  • app_store_reviews.
  • google_play_reviews.
  • amazon_product.
  • product_hunt_search.

Developer workflow

Tool design principles

  • Narrow input schema.
  • Clear description.
  • Bounded result count.
  • Structured output.
  • Visible failure states.
  • Avoid arbitrary user-controlled targets unless intended.
  • Track credit usage.

Developer workflow

Example tool schema

Illustrative schema. Adapt field names to your agent framework.

Tool schema · json

{
  "name": "crawlora_google_search",
  "description": "Search public Google results through Crawlora and return structured JSON.",
  "parameters": {
    "type": "object",
    "properties": {
      "query": { "type": "string" },
      "country": { "type": "string", "default": "us" },
      "language": { "type": "string", "default": "en" },
      "limit": { "type": "number", "default": 10 }
    },
    "required": ["query"]
  }
}

Developer workflow

Example tool function

Keep the Crawlora API key server-side and return explicit failure states to the agent.

Callable tool · typescript

async function crawloraGoogleSearch({
  query,
  country = "us",
  language = "en",
  limit = 10,
}: {
  query: string;
  country?: string;
  language?: string;
  limit?: number;
}) {
  const response = await fetch("https://api.crawlora.net/api/v1/google/search", {
    method: "POST",
    headers: {
      "x-api-key": process.env.CRAWLORA_API_KEY ?? "",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({ keyword: query, country, language, limit, page: 1 }),
  });

  if (!response.ok) {
    return { ok: false, status: response.status, body: await response.text() };
  }

  return { ok: true, data: await response.json() };
}

Developer workflow

Example agent workflow

01

Research request

A user asks for market research on a product category.

02

Search tool

The agent calls Crawlora Google Search for current public search results.

03

Secondary source

The agent calls Product Hunt, app review, or platform-specific endpoints where relevant.

04

Synthesis

The agent summarizes results with source and platform metadata.

Developer workflow

Error handling for agents

  • Invalid input.
  • No results.
  • Rate limit.
  • Upstream unavailable.
  • Challenged or unusable upstream response.
  • Credit limit exceeded.

Developer workflow

Security and responsible use

  • Do not expose the API key.
  • Validate input.
  • Do not allow unrestricted collection abuse.
  • Comply with laws, third-party rights, and platform rules.
  • Avoid sensitive personal data workflows.

Responsible public web data workflows

Use Crawlora for structured public web data workflows. Customers are responsible for compliance with applicable laws, third-party rights, platform rules, and Crawlora terms. Keep API keys server-side, validate inputs, and avoid collecting or storing unnecessary sensitive data.

Read Crawlora terms

Developer workflow

Related developer links

Use these pages to move between endpoint discovery, examples, pricing, and responsible-use guidance.

MCP

AI agents

TypeScript guide

Python guide

Docs

Terms

Developer workflow

FAQ

Common questions for this Crawlora developer integration path.

Can OpenAI Agents call Crawlora APIs?

Yes. Expose Crawlora endpoints as server-side callable tools in your agent workflow.

Should Crawlora be exposed as one generic tool or many narrow tools?

Prefer many narrow tools with clear schemas, bounded result counts, and source-specific descriptions.

How do I keep API keys safe?

Store keys in server environment variables and never include them in browser-side code or model-visible logs.

Can agents use Crawlora for research?

Yes. Crawlora can provide structured public web data for search, local business, app, video, product, and startup research workflows.

Can agents use Crawlora for YouTube transcript summaries?

Yes, when the selected YouTube endpoint supports the transcript workflow you need.

How do I control credit usage?

Set result limits, cache repeated calls, add user quotas, and monitor credits.

What should I avoid?

Avoid exposing arbitrary unrestricted tools, logging secrets, collecting unnecessary sensitive data, or ignoring legal and platform requirements.

Next step

Add Crawlora as a narrow agent tool

Start with one source-specific endpoint, keep the key server-side, and return structured data plus clear failure context.

View AI Agent GuideBrowse APIs