canonical: https://jentic.com/apis/autom.dev/autom # Autom API Reference The Autom API returns structured search engine results from Google, Bing, Brave, DuckDuckGo, and Yahoo as JSON. Google coverage spans web search, Search Light, AI Mode, News, Images, Videos, Shopping, Maps, Jobs, and autocomplete, with location, country, language, and Google-domain targeting. A single `/v1/search` call can walk several engines in order and return the first organic results found. Finder endpoints look up supported Google locations, languages, and countries, and long-running requests such as image searches can run as jobs that are polled for their result. ## For AI agents Fetch structured Google, Bing, Brave, DuckDuckGo, and Yahoo search results, including Google News, Shopping, Maps, Jobs, Images, and AI Mode answers, with location and language targeting. ## Scope Does not crawl full web pages, index your own content, or run paid ad campaigns. Use for fetching structured search engine results only. ## Capabilities - Run a Google web search with location, country (gl), language (hl), and time-range (tbs) targeting and get organic results as JSON - Query Google AI Mode for a question and read its text_blocks, references, and shopping_results - Search Google News, Shopping, Images, and Videos for a query and page through the results - Find places on Google Maps with address, website, and phone, focused around an ll coordinate hint - Pull job listings from Google Jobs with title, company, and location, paging with next_page_token - Fall back across engines in one call: `/v1/search` tries Google, Brave, Bing, DuckDuckGo, Yahoo, and Baidu in order - Look up supported Google locations, languages, and countries before targeting a search ## Use cases ### Web search grounding for an agent through Jentic An AI agent answering a factual question needs fresh search results instead of its training data. Through Jentic the agent searches for 'search Google for a query', receives the Autom Google Search operation with its input schema, and calls GET `/v1/google/search` with the query, location, and hl language. It reads the organic results with titles, links, and snippets and cites them in its answer. Example prompt: Call GET `/v1/google/search` with query 'latest EU AI Act guidance', gl 'gb', and hl 'en', then summarise the top five organic results with their links ### Local rank and visibility checks An SEO consultant checks where a client appears in results for a city. GET `/v1/google/search` accepts a human-readable location such as 'Paris, France' or a uule code, plus google_domain and gl, so the same keyword can be checked from different markets. GET `/v1/finder/google-locations` returns the location strings and IDs Google supports, and GET `/v1/google/maps` returns local place listings with address and phone. Example prompt: Call GET `/v1/finder/google-locations` with query 'Lyon', then GET `/v1/google/search` with query 'plombier lyon', location from the finder result, and google_domain 'google.fr', and report the position of a client domain ### Job and shopping market scans A researcher tracks hiring or product pricing in a market. GET `/v1/google/jobs` returns structured job listings with title, company, and location and pages with next_page_token, while GET `/v1/google/shopping` returns product listings for a query with location and country targeting. Both return JSON that can be stored and compared over time. Example prompt: Call GET `/v1/google/jobs` with query 'data engineer' and location 'Berlin, Germany', follow next_page_token for two more pages, and count listings by company ### Multi-engine fallback search A pipeline that must always return results cannot depend on one engine. GET `/v1/search` walks engines in order, by default Google, Brave, Bing, DuckDuckGo, Yahoo, and Baidu, and returns the first organic results found. The engines parameter sets a custom order, and dedicated endpoints such as GET `/v1/brave/search` and GET `/v1/bing/search` query one engine directly. Example prompt: Call GET `/v1/search` with query 'open source vector database' and engines 'brave,bing,duckduckgo', then list the returned titles and links ## Key endpoints | Method | Path | Description | | --- | --- | --- | | GET | `/v1/google/search` | Google Search results | | GET | `/v1/google/search/ai` | Google AI Mode answer | | GET | `/v1/google/news` | Google News results | | GET | `/v1/google/maps` | Google Maps place results | | GET | `/v1/google/jobs` | Google Jobs listings | | GET | `/v1/search` | Multi-engine fallback search | | GET | `/v1/bing/search` | Bing Search results | | GET | `/v1/jobs/{id}` | Poll an asynchronous job | ## Key resources - **Google search** — Web, Search Light, AI Mode, News, Images, Videos, Shopping, Maps, Jobs, and autocomplete results - **Other engines** — Bing, Brave web, Brave images and videos, DuckDuckGo, and Yahoo results - **Finders** — Supported Google locations, languages, and countries - **Jobs and usage** — Asynchronous job results, account usage and credit balance, and engine health stats ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 57 / 100 - **Maturity:** Foundational - **Dimensions:** - Foundational Compliance: 64 / 100 - Developer Experience & Jentic Compatibility: 68 / 100 - AI-Readiness & Agent Experience: 38 / 100 - Agent Usability: 94 / 100 - Security: 50 / 100 - AI Discoverability: 64 / 100 - **View full report:** https://jentic.com/apis/autom.dev/autom/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring Autom by hand means sending the x-api-key header, learning separate parameter sets for each engine, and handling Google location targeting with location or uule. Through Jentic you install once, import Autom from the API Directory, store the key once, and your agent calls it. - **Permission scoping:** Autom puts the engine and result type in the URL path (`/v1/google/news`, `/v1/bing/search`), so a rule can limit your agent to specific engines and result types. You choose the operations it may call, so ones that cost more credits, like Google AI Mode, are not included unless you add them. - **Credential handling:** Your Autom API key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'search Google for a query' or 'find places on Google Maps', and Jentic returns the matching Autom operation with its input schema so the agent calls the right engine endpoint without browsing the reference docs. ## Related APIs - **SerpApi** — Structured search engine results across Google and many other engines - **Brave Search API** — Brave's own independent web search index - **Tavily API** — Search API that returns condensed results for LLM grounding - **Firecrawl API** — Scrape and crawl web pages into clean markdown ## FAQ ### What authentication does the Autom API use? The Autom API authenticates with an API key in the `x-api-key` request header, declared as the ApiKeyAuth scheme in its OpenAPI spec. Through Jentic the key is held by your own Jentic One instance and attached to each call, so the agent never sees it. ### Which search engines does the Autom API cover? The spec has endpoints for Google (web search, Search Light, AI Mode, News, Images, Videos, Shopping, Maps, Jobs, and autocomplete), Bing, Brave (web, images, and videos), DuckDuckGo, and Yahoo. GET `/v1/search` also walks engines in order and can include Baidu in its default fallback list. ### How are Autom API calls priced? Autom bills in credits. Its spec lists most search endpoints at one credit and Google AI Mode at three credits, while finder endpoints and job polling are free. GET `/v1/usage` returns your current usage and credit balance. ### What are the rate limits for the Autom API? The OpenAPI spec does not specify rate limits. Check the Autom documentation at https://autom.dev for current limits on your plan. ### How do I search Google News through Jentic? Search Jentic for 'get Google News results for a topic', load the GET `/v1/google/news` operation, and execute it with query plus optional gl, hl, and page. The agent receives structured news results it can summarise. To run it on your own infrastructure, install Jentic One from its GitHub repo. ### Is there an Autom MCP server? You don't need an MCP server to give your agent Autom. Jentic connects it directly from the API Directory: import it, store your API key once, and your agent discovers the search operations it needs on demand, so a long list of per-engine tool definitions never has to sit in its context. ### Can I limit what my agent is allowed to do with the Autom API? Yes. If the agent only needs web and news results, write a rule that allows GET `/v1/google/search` and GET `/v1/google/news` and nothing else. It can then run those two searches but not the AI Mode endpoint that costs more credits, and every call it makes is logged.