Skip to content
English
  • There are no suggestions because the search field is empty.

Getting Started with the AI Agent

 

The AI Agent (VIVA) is a multi-step search and procurement assistant built into your Labviva storefront. Instead of scanning for the right keyword, you ask for what you actually need including the best price, a comparison, a protocol and its supplies, or a safety summary and the agent searches the catalog across suppliers, weighs the tradeoffs, and hands back an answer you can act on right away.

This guide shows you how to turn the agent on, what to ask, and how to read what it gives back. It takes less than a minute to get started.

1. Turn the agent on

The agent lives in the search bar at the top of every page. Next to the search box is an AI Agent toggle.

1. Find the AI Agent toggle to the right of the search bar.

2. Click it so it turns bold and is triggered enabled — bold means the agent is on. (Light grey means you're doing a standard keyword search.)

3. Type your request in the search bar and press Enter.

A panel titled VIVA Agent opens over your results. From here the agent works through your request and shows its progress as it goes.

Tip: with the toggle off, the same bar works as a normal search. Flip it on any time you want the agent to assist you.

 

2. What you can ask

The agent is designed around real lab-buying tasks. These are the most common ones — anything phrased naturally will work, and you can fill in your own product, SKU, technique, or chemical where you see the {braces}.

#

What it does

Try asking…

1

Best price search

Find me the best price on {product / SKU}

2

Comparison table

Make me a comparison table of {product / SKU} products, include price per usage

3

Reorder & deals

Check my order history and help me find current deals and better prices for products I've purchased before

4

Protocol → products

Write me a protocol for {product} using {technique} and find me the products I need to do this on Labviva

5

Most-cited product

Find me the most highly cited {product} in literature for {application} that's available on Labviva

6

Spec-based search

I need a {product} with this {specification, e.g. 8-channel pipette} — find me the best price

7

Paper → protocol

Write me a protocol based on this paper {DOI or link} and find me products on Labviva to run it

8

Availability check

Find me the current availability for {product / SKU}

9

Structure & sourcing

Draw me a structure for {chemical} and find the best price with supplier availability on Labviva

10

Safety & compliance

What are the hazards for {chemical} I should be aware of, and how do I store it?

 

The agent can do more than this. These are just some flows to use to get started.

 

3. A worked example

Say you need cell-culture reagent and want the best deal. With the toggle on, type:

“Find me the best price on Trypsin-EDTA 0.25%”

The agent then works through a few steps and returns, in one view:

• Product cards for each matching item image, manufacturer, SKU, the number of supplier offers, price, lead time, and an Add to cart button with a quantity selector.

• A comparison table listing each option with its manufacturer, supplier, price, and availability, with clear status flags (in stock, backorder, or confirm availability).

• Plain-language recommendations a “best price” pick, a “best in-stock value” pick (often noting how widely cited the product is), and volume-vs-cost guidance if a larger pack is a better per-mL deal.

• Suggested Actions one-click buttons allowing quick access to actions such as adding a specific option to your cart, showing larger-volume options, or comparing all options in detail.

You can keep the conversation going in the “Ask the agent about products, offers, history, or actions” box at the bottom; for example, “show me the 500 mL size” or “which ships fastest?” The agent keeps the context from your last question.

 

4. Reading the agent's work

While the agent runs, a status line shows what it's doing for example “Crosswalking query terms to catalog structure,” “Synthesizing the strongest matching evidence,” or “Reconciling tradeoffs across suppliers and products.” When it's done, the status reads Ready.

Each answer includes an expandable summary of the tools it used (e.g. “Picked the right tools · 1 tool call made”) so you can see how it reached its result. You don't need to open this to use the answer it's there for transparency.

 

Chat history

Your past agent conversations are saved in the left sidebar of the panel, labeled by date. Click any one to reopen it, or click + New chat to start fresh.

 

5. Tips for better results

• Be specific about what matters. “Best price,” “fastest shipping,” “most cited,” “largest pack” all steer the answer.

• Include specs when they count. Concentration, volume, channel count, sterility, animal-origin; the more precise, the tighter the match.

• Use a SKU or catalog number when you know exactly what you want; use a description when you're exploring.

• Ask follow-ups. You don't have to get it perfect in one line you can refine in the chat box.

• Sort by citations (or just ask for “most cited”) when reproducibility matters, many catalog items show their publication count.

 

6. Good to know (Beta guidance and quick support)

The AI Agent is in beta and improving quickly. A couple of things to expect:

• If a step fails. You may occasionally see “Agent request failed. please try again.” Just resend your request; it usually goes through on retry.

• Always confirm critical details. Prices, availability, and safety/compliance information should be verified against the supplier offer and the product's SDS before you rely on them for a purchase or handling decision.

• Your feedback helps. Tell us what worked and what didn't! it directly shapes what the agent learns to do next.