Agents that start from what AI answers about your market
They write the pages you're missing and prepare pitches for publications, Reddit replies and your site's structured data. They use your brand's facts and wait for your team's approval.
Want to see how AI engines see you today first? Get the free analysis.

They start from the data
Every agent works on an Action: the prompts where AI doesn't mention you and the sources it cites instead of you.
They write with your facts
Refinea Brain holds your brand's numbers and cases, each with its source, along with your voice and scope.
They wait for your OK
They stop at the outline, the draft or the pitch until your team approves.
They work off your site too
Publications, Reddit and YouTube, as well as content and technical SEO.
Five areas of work, each with its own agents.
Ask Ref for what you need, in your own words. It picks the right agents and shows you the result.
Writes the pages you're missing and reworks the weak ones.
It starts from the questions in your market that none of your pages answer today, writes with your brand's facts and tone, and stops at the outline until you approve it.
A membrane keeps out water that comes from below and from the sides. It doesn't stop water that gets in at the ankle, and on a dry trail it makes your foot warmer than a breathable upper. To choose, start with three questions: where you run, for how long and in what temperatures.
Mud, wet snow, shallow stream crossings and runs of more than an hour in the rain: this is where a shoe with a GORE-TEX membrane keeps your foot dry longer. In the Nike range, that's the GORE-TEX version of the Pegasus Trail 5.
On packed trails, in temperatures above 10°C or in a race, your foot sweats more than it gets wet. A breathable upper dries quickly, even after a stream crossing.
Does water still get in at the ankle? Yes, if it's higher than the shoe. A low gaiter cuts it down.
Agents for every area, ready to use from day one.
Each one does a specific job. You'll find them grouped by category and can launch them from an Action, from a chat with Ref or straight from the library.

- Create Articlewrites the full article from the prompts you monitor
- Content Brief Creationthe outline for your writers
- AEO FAQ Generatorthe FAQs for a page
- PLP AEO Copywriterthe copy for category pages
- Article Translator and Multilanguage Article Translatorup to three languages at once
- YouTube to Blog Postfrom video to article
- Citation Outreachfor every page cited instead of you: what to offer, who to write to and the email draft
- Reddit AI-Citation Strategistreplies to the threads the engines cite
- YouTube Metadata Optimizertitle, description and chapters
- Structured Data GeneratorJSON-LD ready to paste
- Internal Linkerinternal links to the right pages, picked from the sitemap
- Robots.txt & LLMs.txt Auditorchecks that AI engines can find, read and cite the site
- Indexing Bingsubmits pages with IndexNow
- PDP ChatGPT Shopping Auditchecks product pages against the criteria ChatGPT Shopping uses to rank products
Need an agent that doesn't exist? Ask Ref.
Tell it what you need. Ref asks the questions that change the result, then builds the agent on the canvas, node by node, and you see it before you run it.

Ref's questions
Which topics, where to publish, when to stop for approval.
The canvas
Every step is a node you can see and change by hand.
Refinea Brain and Skills
Humanizer and your own Skills go into every node that writes.
From your team's OK to the published page.
The agent stops at the outline or the draft. You review it, ask for a change or approve.

With WordPress and Webflow the article arrives as a draft in the right section, with an SEO title and image. On GitHub it arrives as a pull request with a preview. With other CMSs it arrives ready to paste.

On a sheet, each row is a page and each column is an agent: you see what's done, what's in progress and what's left.

Schedule an agent and it runs again on its own, with fresh data.

“A great platform for AI Search visibility, support that's always there, and a company that really listens to what could be better.”
Christian CilliFounder of Seoleader
Put agents to work on everything AI reads about you.
With Pro, we set up Refinea with you on day one. Agents prepare the work, and your team approves it before anything goes out.
Want to see how AI engines see you today first? Get the free analysis.
Frequently asked questions about agents
Do the agents publish on their own?
No. They stop at the outline, the draft or the pitch until your team approves. With WordPress and Webflow, even after the OK the work lands on your site as a draft; on GitHub, as a pull request.
How is this different from ChatGPT or Claude?
The agents start from the answers AI engines give every day about your market: which prompts you lose and which sources they cite. They write with the facts, voice and scope of your brand saved in Refinea Brain, and they also work off your site, on publications, Reddit and YouTube.
Can I build my own agent?
Yes. Ask Ref in the chat: it asks the questions it needs and builds the agent on the canvas.
Do they work in other languages?
Yes. The agents write in the language of the market you monitor, and Multilanguage Article Translator translates an article into three languages at once.
Which models do they run on?
The ones you choose: you can use all available models and customize every agent, step by step.
How many agents can I run on my plan?
Agents use your plan's credits: 500 a month with Lite, 1,500 with Pro. An article costs 70 credits, an optimization about 30: with Lite that's about 7 articles or 16 optimizations a month.
