Web search now available in the AI agent action

Your EveryCatch AI agent can now search the internet while it runs. Turn on Web search in the new Built-in tools section so the agent can look up the facts it needs in the middle of a workflow. You can also point it to specific pages by pasting their URLs in the Instructions, and the agent will read those pages when it needs them.

What changed

A built-in tools section

• The Built-in tools section now appears right below Tools in the AI agent action. These are capabilities your agent has automatically, in addition to the tools you choose to add yourself.
• Hover over the (i) icon on any built-in tool to see what it does. Hover over the crown icon to see what it costs.

Web search

• Single switch: Turn on Web search and the agent can search the internet during any run in that workflow.
• Smart usage: The agent decides when to search, and only does it when the task requires information it does not already have.
• Use your own sources: Paste a URL into the Instructions, and the agent will refer to that page when a run needs the information on it.
• Mix static and live sources: Give the agent the trusted pages you want it to rely on, and let it use web search for everything else.

Coming soon to built-in tools

AI image generation and AI video generation are already available as separate actions in EveryCatch. Both features are coming soon into the AI agent action as built-in tools, so you can trigger them directly from a single agent setup.

Pricing

Web search is a premium built-in tool. It is billed per search, in addition to the cost of running the AI agent action.

• OpenAI and Claude models: $0.015 per search.
• Gemini models: $0.021 per search.

How to use

1. Go to Automation > Workflows.
2. Open a workflow, then add a new AI agent action or open an existing one.
3. Scroll down to the Built-in tools section and expand it.
4. Turn on Web search.
5. If you want the agent to use specific pages, paste their URLs into the Instructions field with clear guidance on when to use them. This step is optional.
6. Save the action.

Worked example: rebate questions, answered from the source

A heating and cooling company uses EveryCatch to follow up on every quote request. Some leads ask about rebates, and rebate programs change frequently.

The AI agent action is configured like this

1. Instructions: "If the lead asks about rebates, read https://www.energystar.gov/rebate-finder before you reply. For any other facts you need, search the web."
2. Tools: Send email and Send SMS.
3. Built-in tools: Web search turned on.

A lead asks about heat pump rebates. The agent first checks the rebate page you specified, then replies using the current details. Another lead asks a question that the rebate page does not cover, so the agent performs a web search to find a reliable answer. This way, your team avoids outdated information, guesswork, and manual research.

Why this matters

Your agent no longer depends only on what is in its existing knowledge or prompts. Prices, rebate programs, and public information change regularly, and web search lets the agent check them at the moment it needs them.

You stay in control of where the agent looks. List the pages you trust in the Instructions and tell the agent when it should consult them.

You are charged only when the agent performs a search. A run that completes without searching the web does not add any web search fees.

Good to know

• Each search is billed separately. If a single run triggers three searches, you are charged for three searches.
• Web search helps the agent find information. To follow up on what it finds, the agent still needs the appropriate tools, such as Send email or Send SMS, added to the action.