canonical: https://jentic.com/apis/linkup.com/linkup # Linkup Search API The Linkup API runs real-time web searches and returns source-backed answers to natural-language queries. It searches the web, fetches the contents of a single page, generates AI responses grounded in live search context, runs longer asynchronous research tasks, and reports your remaining credit balance. ## For AI agents Search the live web with source-backed answers, fetch page contents, and run asynchronous research tasks for an AI agent. ## Scope Does not index your private documents, store search history, or crawl entire domains. Use for real-time web search and page fetching only. ## Capabilities - Search the live web and return source-backed answers to a query - Fetch the full contents of a single webpage - Generate an AI response grounded in fresh web-search context - Create and poll an asynchronous research task - Read your remaining credits balance ## Use cases ### Grounded answers for an AI agent through Jentic An AI agent answering current-events questions can search the live web and generate a source-backed response. Through Jentic the agent discovers the search and response operations by intent and loads their schemas, so it grounds answers in fresh results without wiring the Linkup endpoints. Example prompt: Search Jentic for 'search the web', run a query for a topic, and generate a source-backed answer from the results. ### Single-page content retrieval Fetch the full contents of a specific page when the agent already has the URL, rather than running a broad search. Linkup returns the page content so the agent can read and reason over it directly. Example prompt: Fetch the full content of a known URL and extract the section relevant to the user's question. ### Long-running research tasks Kick off an asynchronous research task for a broad question and poll it until it finishes, then read the compiled result. Linkup runs the task in the background so an agent does not block on a long query, and the credits balance lets it pace usage. Example prompt: Create a research task for a broad question, poll it until complete, then read the compiled result and check the remaining credits. ## Key endpoints | Method | Path | Description | | --- | --- | --- | | POST | `/v1/search` | Search the web for a query | | POST | `/v1/fetch` | Fetch content from a webpage | | POST | `/v1/responses` | Generate an AI response using web search | | POST | `/v1/research` | Create an asynchronous research task | | GET | `/v1/credits/balance` | Retrieve current credits balance | ## Key resources - **Searches** — Web searches and source-backed answers - **Research tasks** — Asynchronous research jobs and their results - **Credits** — Remaining credit balance for the account ## Why Jentic - **Setup:** Wiring Linkup by hand means sending its bearer key, choosing between search, fetch, and research operations, and polling asynchronous tasks yourself. Through Jentic you install once, import Linkup from the API Directory, store the key once, and your agent calls it. - **Permission scoping:** Linkup carries the query in the request body, so rules bound which operations your agent may call, not which sites. You choose the operations it may call, such as search and fetch, so starting a research task is not included unless you add it. - **Credential handling:** Your Linkup 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 the web' or 'fetch a webpage', and Jentic returns the matching Linkup operation with its input schema so the agent calls the right endpoint without browsing the reference docs. ## Related APIs - **Tavily** — Search API built for AI agents with grounded answers - **Brave Search** — Independent web search index with an API - **Exa** — Neural search over the web for retrieval ## FAQ ### What authentication does the Linkup Search API use? The Linkup Search API authenticates with a bearer token per its OpenAPI spec: pass your API key as a Bearer token in the Authorization header. Through Jentic the key is stored encrypted by your own Jentic One instance and injected at request time. ### Is there a Linkup Search API MCP server? You don't need an MCP server to give your agent the Linkup Search API. Jentic connects it directly from the API Directory: import it, store your credential once, and your agent calls it on demand without loading another server's tool definitions into its context. ### Can I get answers with sources from the Linkup Search API? Yes. The search operation returns source-backed results, and the response operation generates an AI answer grounded in live web-search context, so an agent can cite where each claim came from. ### What are the rate limits for the Linkup Search API? The OpenAPI spec does not state numeric rate limits; usage is metered against a credits balance the API reports. Check the Linkup documentation at https://docs.linkup.so for current limits and pricing. ### Can I limit what my agent is allowed to do with the Linkup Search API? Yes. Linkup carries the query in the request body, so rules bound which operations your agent may call. Allow only the search and fetch operations it needs, so it can look things up but cannot start expensive research tasks unless you add that operation. ### How do I search the web with the Linkup Search API through Jentic? Search Jentic for 'search the web', load the search operation, and pass your query. Jentic injects your key and returns source-backed results. To run it on your own infrastructure, install Jentic One from its GitHub repo.