How does the subscription work?

It is a simple monthly "Seat" license. Because we cover the expensive AI computing costs (API tokens) for every file you process, we charge a flat monthly fee rather than a large upfront cost. This ensures you always have access to the latest, smartest AI models without ever seeing a surprise bill from OpenAI.

What software do I need to run these agents?

You need a free or paid instance of n8n. All our workflows are built on n8n, the world's most powerful workflow automation platform. You can run this on their cloud service (starts at ~$20/mo) or, if you are technical, you can self-host the open-source version on your own server for free.

Can I share my subscription with my agency?

No. The API Key is locked to one user. Your subscription grants you a unique, encrypted API Key that authenticates your requests. This key is monitored for abnormal volume and concurrent usage. If you need to deploy this for an entire Research Team, please contact us for our Enterprise package, which allows for shared billing and higher processing limits.

What is the Cancellation Policy?

You can cancel anytime. There are no long-term contracts. If you cancel, your access to the Intelligence Engine will stop at the end of your current billing cycle, but you get to keep every slide deck you have generated forever.

How is my data handled? Is it safe?

Yes. We use "Ephemeral Processing." Your transcripts are sent to our secure processing engine solely for the purpose of generating your slides. We do not store, train on, or archive your user data. Once the slide deck is generated and returned to you, the raw data is purged from our processing cache. Your secrets stay yours.

Do you use my proprietary data to train your AI models?

Absolutely not. This is a hard line for us. We utilize "Zero-Retention" API protocols with our AI partners. Your data is sent to the reasoning engine, processed into a slide deck, and immediately discarded from the inference cache. We do not use your interviews to train our models, and we do not allow OpenAI or Anthropic to use them either.

Which AI model does this engine run on?

We use a "Best-in-Class" orchestration strategy. We don't just default to the newest model; we conduct extensive benchmarking to find the perfect LLM for each specific agent. For example, our research proves that Claude 4.5 Sonnet currently outperforms GPT-4o in strategic nuance and qualitative synthesis, so that is what powers the "Boardroom Researcher." As a subscriber, you benefit from our ongoing R&D: if a new model emerges that performs this specific task better, we swap the engine instantly, ensuring you are always running on the optimal brain for the job.

Is there a limit on how many outputs I can generate?

Yes, we adhere to a "Fair Use" policy. Accounts are allocated a generous monthly allowance (typically 15-20 deep-dive studies). High-volume users can upgrade to Enterprise Plans.

What happens if the agent fails or "hallucinates" on a file?

We monitor the engine 24/7. Because the logic runs on our managed infrastructure, we can see if a job fails due to a system error. If a run fails on our end, it does not count toward your usage limit, and we will often re-run it for you automatically. If you encounter a specific issue with a file, our support team is available to help debug the input.

Do I need to reinstall anything when you update the Agent?

No. Updates are automatic. This is the beauty of the hosted subscription. When we improve the "Senior Researcher" prompt or upgrade the underlying AI model, your "Connector Node" automatically benefits from those changes instantly. You don't need to download new files or patch your workflow—it just gets smarter overnight.