canonical: https://jentic.com/apis/andiai.com/andiai # Andi AI Search API The Andi AI Search API searches the web, fetches and extracts single pages, and serves curated news feeds for people and AI agents. Its search operation returns structured results with query operators, domain and date filtering, and multiple output formats, including a context format that returns markdown suited to language models. The fetch operation retrieves and extracts one page, and the news operation returns a ranked feed for one of 26 fixed topics with no query required. Results are built for retrieval, so an agent can ground its answers in current web content. ## For AI agents Search the web with query operators and domain and date filters, fetch and extract a single page, and pull a ranked news feed for one of 26 fixed topics. A context output format returns markdown suited to grounding language models. ## Scope Does not crawl entire sites, store a persistent index, or generate answers itself. Use for web search, single-page fetch, and curated topic news only. ## Capabilities - Search the web with query operators and domain and date filtering, via GET or a JSON POST body - Return results in multiple output formats, including a context format of markdown for language models - Fetch and extract the content of a single web page - Pull a curated, ranked news feed for one of 26 fixed topics with no query required - Ground an agent's answers in current web content through retrieval-ready results ## Use cases ### Grounding an Agent in Live Web Results An agent answering a question that needs current information can run a search and request the context output format, which returns markdown shaped for language models, then feed those results straight into its prompt. Query operators plus domain and date filtering let the agent narrow to trustworthy, recent sources before grounding. This gives a retrieval-augmented answer without the agent operating its own crawler or index. Example prompt: Search for a question with a date filter for the last month, request the context format, and pass the returned markdown results to the model as grounding. ### Fetching a Specific Page When an agent already has a URL, the fetch operation retrieves and extracts the single page's content so the agent reads clean text instead of raw HTML. This suits following a citation, reading a documentation page, or extracting an article the agent found through search. One call turns a link into usable content. Example prompt: Fetch a specific article URL, extract its main content, and summarise it for the user. ### Curated Topic News Feeds The news operation returns a ranked feed for one of 26 fixed topics without requiring a query, so an agent can surface the latest items in a subject area in a single call. Because the topics are fixed and the feed is curated and ranked, the agent gets a stable, low-effort way to monitor a domain. It suits briefings and recurring digests. Example prompt: Pull the ranked news feed for a chosen fixed topic and compile the top items into a short briefing. ## Key endpoints | Method | Path | Description | | --- | --- | --- | | GET | `/api/v1/search` | Search the web with query operators and filters | | POST | `/api/v1/search` | Search the web with a JSON request body | | GET | `/api/v1/fetch` | Fetch and extract a single web page | | GET | `/api/v1/news/{topic}` | Get a curated, ranked news feed for a fixed topic | ## Key resources - **Search** — Search the web via GET or a JSON POST body with query operators, domain and date filters, and multiple output formats - **Fetch** — Retrieve and extract the content of a single web page - **News** — A curated, ranked news feed for one of 26 fixed topics, returned as JSON ## 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:** 69 / 100 - **Maturity:** AI-Aware - **Dimensions:** - Foundational Compliance: 93 / 100 - Developer Experience & Jentic Compatibility: 70 / 100 - AI-Readiness & Agent Experience: 53 / 100 - Agent Usability: 94 / 100 - Security: 50 / 100 - AI Discoverability: 84 / 100 - **View full report:** https://jentic.com/apis/andiai.com/andiai/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 the Andi AI Search API by hand means creating a key in the API Console and sending it in the x-api-key header on search, fetch, and news calls, and choosing the right output format for each. Through Jentic you install once, import it from the API Directory, store the key once, and your agent searches, fetches, and reads news. - **Permission scoping:** A rule can allow your agent only the operations it needs, such as search and page fetch, so it can read the web while you leave out whatever you do not want. The news feed takes the topic in the URL path (`/api/v1/news/{topic}`), so you can allow that read as its own operation, and rules bound which operations run. - **Credential handling:** Your Andi 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 page as markdown', and Jentic returns the matching Andi operation with its input schema so the agent calls the right endpoint without browsing the reference docs. ## Related APIs - **Tavily** — Web search built for retrieval-augmented generation, like Andi's context-format results - **Exa** — Neural and keyword web search with content retrieval rather than fixed-topic news feeds - **Brave Search** — Independent web search index returning results and snippets without a page-fetch or news endpoint ## FAQ ### What authentication does the Andi AI Search API use? Per its OpenAPI spec the Andi AI Search API uses an API key sent in the x-api-key header, which you create in the API Console. The key applies to the search, fetch, and news operations alike, so every call carries it in that header. ### Can I get search results formatted for a language model with the Andi AI Search API? Yes. The search and fetch operations support a context output format that returns markdown suited to language models, so you can pass results straight into a prompt. Search also supports query operators and domain and date filtering to narrow to the sources you want before grounding. ### What are the rate limits for the Andi AI Search API? The OpenAPI spec does not publish numeric rate limits for the Andi AI Search API. For current limits check the API Console and vendor documentation at https://console.andiai.com. ### How do I run a web search with the Andi AI Search API through Jentic? Import the Andi AI Search API from the Jentic API Directory, then have your agent search by intent such as 'search the web for recent results' and Jentic returns the search operation with its input schema. The agent runs the query, requests the context format, and grounds its answer in the results. To run it on your own infrastructure, install Jentic One from its GitHub repo. ### Can I limit what my agent is allowed to do with the Andi AI Search API? Yes. Write a rule that allows only the operations the agent needs, such as search and page fetch, so it can read the web but you leave out the news feed if it is not wanted. The news feed takes the topic in the URL path, so you can allow that read as its own operation, and every call the agent makes is logged. ### Is there an Andi AI Search MCP server? You don't need an MCP server to give your agent the Andi AI Search API. Jentic connects it directly from the API Directory: import it, store your API key once, and your agent calls the search, fetch, and news operations on demand without loading another server's tool definitions into its context. ### What topics does the news feed cover in the Andi AI Search API? The news operation returns a curated, ranked feed for one of 26 fixed topics, selected by naming the topic in the path, and it requires no search query. The feed is returned as JSON rather than the context markdown format offered by search and fetch.