For Agents
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the goog.io | Unofficial Google Search API, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
# On the machine that will host your Jentic One instance:
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
What an agent can do with goog.io | Unofficial Google Search API.
Run a Google web search and return ranked organic results
Search Google Images with country and language localisation
Fetch Google News results for a query and time window
Crawl a target query through goog.io's USA-IP infrastructure
GET STARTED
Run Google web, image, news, crawl, and SERP queries via 6 endpoints with country and language targeting on goog.io.
Use for: Search Google for the top 10 organic results for 'agentic AI startups', I want to retrieve Google News articles about a competitor in the last 24 hours, Get a full SERP for a target keyword to feed an SEO report, Find Google Images for a product name and return their URLs
Not supported: Does not handle JavaScript page rendering, indexing your own site in Google, or paid ad placement — use for unofficial Google web, image, news, crawl, and SERP retrieval only.
Jentic publishes the only available OpenAPI specification for goog.io | Unofficial Google Search API, keeping it validated and agent-ready. The goog.io API exposes 6 endpoints for running Google web search, image search, news search, page crawling, and SERP retrieval programmatically without managing proxies. Endpoints accept a query as a path parameter and return clean structured results, with optional language, country, and locality parameters for geographic targeting. Authentication is an apikey header.
Retrieve a full Google SERP for ranking and SEO research
Check service status before issuing search jobs
Patterns agents use goog.io | Unofficial Google Search API for, with concrete tasks.
★ SEO Rank Tracking
SEO teams pull Google SERPs for tracked keywords daily and write the results into a dashboard. POST /v1/serp/ returns the structured SERP, and the country and language parameters let teams track rank changes per market in one job.
Call POST /v1/serp/ for the keyword 'best crm api' in the US and UK markets and store the top 10 organic positions.
News Monitoring
Brand and competitive teams call /v1/news/{query} every hour for tracked terms, dedupe results, and feed them into a Slack channel. The infrastructure handles the proxy and IP rotation problem that makes do-it-yourself Google scraping fragile.
Pull /v1/news/competitor-name every 30 minutes and post any article from the last hour into the #news channel.
Lead Research Crawling
Sales and BD teams crawl Google for prospect-related queries and feed the result into an enrichment pipeline. /v1/crawl/{query} returns clean USA-IP results, which avoids the captchas and IP bans that DIY scrapers hit at scale.
For each prospect company in the CRM, crawl '{company} pricing' and store the top 5 result URLs.
AI Research Agent via Jentic
An AI research agent uses goog.io as its 'fresh web' tool through Jentic. The agent searches Jentic for a Google search operation, loads the schema, and executes — Jentic injects the apikey header so the agent only handles the query, country, and language.
Search Google for the user's question, take the top 5 organic URLs, and pass them to a downstream summarisation step.
6 endpoints — jentic publishes the only available openapi specification for goog.
METHOD
PATH
DESCRIPTION
/
Service status
/v1/search/{query}
Google web search
/v1/images/{query}
Google Images search
/v1/news/{query}
Google News search
/v1/crawl/{query}
Google crawl
/v1/serp/
Full Google SERP
/
Service status
/v1/search/{query}
Google web search
/v1/images/{query}
Google Images search
/v1/news/{query}
Google News search
/v1/crawl/{query}
Google crawl
/v1/serp/
Three things that make agents converge on Jentic-routed access.
Credential isolation
goog.io apikey values are stored encrypted in the Jentic vault (MAXsystem) and scoped to a single subscription. Agents receive a short-lived execution handle, so the raw apikey never enters the model context or logs.
Intent-based discovery
Agents search Jentic by intent (e.g. 'google web search') and Jentic returns the matching /v1/search/{query} or /v1/serp/ operation with its parameter schema.
Time to first call
Direct goog.io integration: a few hours of HTTP plumbing and rate-limit handling. Through Jentic: under 15 minutes — search, load, execute.
Alternatives and complements available in the Jentic catalogue.
Specific to using goog.io | Unofficial Google Search API through Jentic.
Why is there no official OpenAPI spec for goog.io | Unofficial Google Search API?
goog.io does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call goog.io | Unofficial Google Search API via structured tooling. It is validated against the live API and kept up to date. Get started at https://app.jentic.com/sign-up.
What authentication does the goog.io API use?
goog.io uses an apiKey scheme on the apikey header. Through Jentic the key is stored encrypted in the vault and scoped to your goog.io subscription, so the raw key never enters the agent's context.
Can I run a Google SERP query with goog.io?
Yes. POST /v1/serp/ accepts a query and optional country, language, and locality parameters and returns the structured SERP, including organic positions and any rich result blocks Google rendered.
What are the rate limits for the goog.io API?
goog.io enforces per-subscription rate limits visible in your account dashboard. The OpenAPI spec does not encode exact numbers, so check the dashboard before running large rank-tracking jobs — pace concurrent requests rather than firing them in tight loops.
How do I run a Google search through Jentic?
Search Jentic for 'google web search', load the schema for GET /v1/search/{query}, and execute with the encoded query in the path. Jentic injects the apikey header so the agent only supplies the query and any localisation parameters.
Does the API return JavaScript-rendered content?
No. The endpoints return Google's results pages and crawled URLs as text or JSON; they do not execute JavaScript on the destination pages. For full-page rendering, fetch each result URL with a headless browser API.
Full Google SERP