Answers a question with citations, extracts full page text, builds Websets of lookalike pages, and runs async research tasks that return reports.
18actions available
Every action runs live through MCP. Nothing to build, nothing to maintain.
Ayro launches an async research task that explores the web, gathers sources, and synthesizes findings with citations. Hands you a comprehensive brief.
Ayro synthesizes a direct, citation-backed answer to your question by reading relevant pages. Saves you from manual link-clicking and skimming.
Ayro runs a web search with advanced filtering and AI-optimized prompting. Returns relevant results faster than raw search engine queries.
Each card is one request a support team gets, and the Exa actions Ayro runs to close it.
Ayro generates a direct answer with its citation list, or runs a filtered search and then pulls the full text and highlights from the URLs it returns, since search alone gives only snippets.
From a single URL, Ayro finds semantically similar pages using embeddings, then seeds a Webset with that search and uploads whatever data you already hold into the same collection.
Ayro starts an asynchronous research task that gathers sources and synthesises findings, takes the researchId back immediately, and polls until the report or structured JSON is ready.
Ayro schedules a Monitor so a Webset updates without a manual run, then reads the webhook registrations and the event history to confirm those updates actually fired.
How it works
Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Ayro reads the Exa actions it has and picks the ones a request needs.
Authorize the Exa account your team already uses — one consent screen, no API keys, no mapping tables. Ayro can only do what you granted that account, and you can disconnect it at any time.
Describe the job the way you would hand it to a teammate. Ayro maps it to the Exa actions that close it and chains as many as the request needs.
The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.
Get started freeExa · Ayro
What can you do in Exa without me setting anything up?
These, right away:
Great, let's start with the first one.
FAQ
Setup, permissions, and the limits of what Ayro can do inside Exa.
Two that matter. includeDomains and excludeDomains are mutually exclusive and sending both is a validation error, and type='keyword' supports neither of them nor useAutoprompt - only 'neural' and 'auto' do. Ayro picks the search type to match the filters a question actually needs.
All 18 actions your agent can call on Exa, straight from the live MCP connection.
Generates a direct, citation-backed answer to a clear natural language question or topic using Exa's search, adept at both specific answers and detailed summaries for open-ended queries.
Create a new import to upload data into a Webset.
Create a new Monitor.
Creates an asynchronous research task that explores the web, gathers sources, synthesizes findings, and returns results with citations.
Create a new Webset with search, import, and enrichment setup.
Delete an existing import.
Delete a Webset.
Finds web pages semantically similar to a given URL using embeddings-based search, optionally retrieving full text, highlights, or summaries for results.
Retrieve details of a specific API key by its ID.
Retrieves configurable text and highlights from a list of Exa document IDs or publicly accessible URLs.
Get details of a specific event by its ID.
Retrieve the status and results of a previously created research task.
List all events that have occurred in the system.
List all imports for the Webset.
Retrieve a paginated list of your research tasks.
List all webhooks for Websets.
Performs a web search using the Exa engine, useful for queries requiring advanced filtering, specific content categories, or AI-optimized prompting.
Update an import configuration by ID.
Same connection, same setup. Pick the next one your team already uses.
Runs a trained robot on demand or 1,000 tasks in one call, schedules change monitors, and reports what each extraction returned.
Runs a filtered statistics query against FaceUp reporting data so activity questions get a real figure in the conversation.
Reads the live data behind a dashboard, forces a stale data source to refresh, and manages the users, roles, and share rights around it.
Starts schema-driven research runs, builds and enriches FindAll entity lists, batches runs into task groups, and monitors queries on a cadence.
Records conversation events, updates customer traits, links anonymous visitors to known profiles, and reports source and destination delivery health.
Pulls live and historical WHOIS records, finds every domain tied to a company or email, enumerates subdomains, and scores an IP for threat activity.

Authorize the account and your agent has all 18 actions from the first conversation.
