Frequently asked questions

RaLHF is a context assistant, not another chatbot. It builds a private library of your context, everything that makes you you, and serves the right slice to whatever AI you use so it stops guessing. It works with Claude, Codex, and ChatGPT today.

A context assistant has one job: manage your personal context and feed it to the AI you already use. It does not answer your questions itself. It makes the AI you choose answer them better, because that AI finally knows who you are and has access to your personal context library.

In big companies, context engineers hand-build the background an AI needs to do a job well. RaLHF does that same work for you automatically: collecting your context, keeping it current, organizing it, and serving the right pieces to your AI for each task.

Neither. RaLHF is the context layer that sits between you and the AIs you use. It does not replace Claude, Codex, or ChatGPT. It gives them the personal context they are missing, so you get answers built around your real life and work.

ChatGPT and Claude are the AIs you talk to. RaLHF is not a competitor to them. It feeds them your context so their answers fit you. You keep using the AI you like, and RaLHF makes all of them more personal.

Because it does not know you. An AI knows almost everything published online and almost nothing about your preferences, history, or work. Without that context it fills the gaps with confident guesses, so you get generic answers to personal questions every time.

Context rot is the slow buildup of stale, contradictory details in an AI’s memory that quietly drags down the quality of its answers. The longer a chat runs, the worse it gets. RaLHF avoids it by serving fresh, relevant context for each task instead of piling everything up.

Because most AI has limited memory and only stores limited information from your past conversations. Your brand, your preferences, the project you set up last week, not stored for the long term. RaLHF holds that context for you, keeps it organized and hands it to the AI automatically, so you stop pasting the same background in every conversation.

Only a little. Built-in memory keeps a few facts and recent chats, tied to that one app, and the AI company decides what it keeps. It does not know your wider digital life and it does not travel to other AIs. RaLHF does both, and you own it.

You connect your apps and upload anything that matters. RaLHF reads it, keeps it current, and organizes it into a context library you own. When you work with an AI, it serves the right slice for that task, so the AI answers with your context in hand.

MCP, the Model Context Protocol, is an open standard that lets AI assistants connect to outside sources like RaLHF. It is how your chosen AI securely reaches the context you have approved. Any AI that supports MCP can connect, and more add support all the time.

Connect web apps and services, upload documents like PDFs, work documents and health records, forward email to RaLHF (feed@RaLHF.ai), or add notes by hand. The more you add, the more personal and effective every AI task becomes. RaLHF then keeps each source current for you.

Yes. Once you connect a source, RaLHF refreshes it automatically at the right pace, so your context stays current without re-uploading. You set it up once, and it keeps working in the background.

No, RaLHF prepares your context ahead of time, so when your AI needs it, the right slice is already organized and ready to serve. The AI is not digging through your whole library on every question. You get answers that fit you, at the same speed you are used to.

Yes, that is the point. You build your context library once, then use it with Claude, Codex, ChatGPT, and any AI that connects. You do not rebuild it for each tool. One library, every AI you work in.

Yes. Your context library belongs to you, not to any AI, so it follows you when you switch. Move from ChatGPT to Claude or add Codex tomorrow and everything you have built comes with you. No starting over.

No. You set it up once in RaLHF. Connecting a new AI takes a moment and it immediately works from the context you already have. There is no separate setup, no re-uploading, and no copying answers between tools.

In the app, open your AI apps, pick the assistant you want, and follow the one-click connection. It uses MCP behind the scenes. After that, the AI you chose can use the context you have approved, automatically.

Yes. You grant access per AI. Give Claude your work and travel, let ChatGPT see entertainment, keep finances private from everything. Each AI only sees what you allow, and you can change it whenever you want.

Yes. Access is granted and revoked with a tap, per AI. Turn an assistant off and it can no longer reach your context. You stay in control of which AIs see your library at all times.

It works with Claude, Codex, and ChatGPT today, with more on the way, plus any AI that supports MCP, the open connection standard. RaLHF is deliberately neutral, so you are never locked to a single AI vendor.

Yes, with shared context libraries. You invite the people you choose, and everyone’s AI can work from the same shared context. It is built for teams, families, and friends who plan or work together.

You create a library, add the context, and invite people to it. Their AI can then draw on that shared context for tasks, while your personal library stays separate. You decide who is in, and you can remove anyone at any time.

Yes. Drop the brief, the account history, and the decisions into a shared library, and every teammate’s AI works from the same source. No forwarding, no wondering who has the latest version, no copy-paste handoffs between people.

Yes. Share a library with family or friends and everyone’s AI can plan around the whole group: the trip everyone will enjoy, the dinner that fits every diet, the gift list that makes sense. One shared context, better group decisions.

No. Only the AI sees the combined context. The people in a shared library do not get access to your raw files or personal data. They get better AI help from the shared context, not a window into your accounts.

Yes. You invite who you want, see exactly who is in, and remove anyone whenever you choose. Sharing is always by invitation and always reversible, so you stay in control of your context.

Yes. Shared libraries are invite-only, encrypted, and limited to the context you choose to include. Nothing is shared without your approval, only the AI reads the combined context, and you can revoke access at any time.

With RaLHF, that is the whole design. Your context is encrypted and stored separately from any AI vendor, never sold, and never used to train models. You decide which AIs see what, and you can export or delete everything at any time.

You do, fully. Your context is yours, not RaLHF’s and not an AI company’s. You control who can use it, you can take a full export whenever you like, and if you delete your account, your data goes with it.

Never. RaLHF does not sell your data, does not use it to train any AI model, and does not show ads based on it. The business runs on a simple subscription, so you are the customer, not the product.

Yes. You can export your entire context library at any time, and delete your account and its data whenever you want. There is no lock-in. Your context is portable by design, so it always stays yours.

Yes. Nothing is shared until you grant access, and every source and every AI is off until you turn it on. Privacy is the default state, not a setting you have to go find and switch on.

RaLHF is built by Bot Food Corporation in Toronto, Canada, and operates under Canadian privacy law, among the strongest in the world. That standard was the baseline the product was built on, not an afterthought.

Built-in memory only remembers your chats with that one app. It does not know your Netflix history, your documents, or your real life, it is locked to that platform, and the AI company owns it. RaLHF goes deeper, works across AIs, and belongs to you.

With built-in memory, you start over from zero. Everything the old AI knew stays locked in its platform. With RaLHF your context library moves with you, so a new AI is useful from the first message.

They are building it, as walled gardens that lock your context inside their platform. RaLHF is the independent alternative: it works across all of them and stays owned by you. The point is a neutral home for your context before the lock-in sets in.

One plan, no tiers. Free for the first 90 days with no credit card, then $5 a month. That includes connecting your apps, uploading anything, support for Claude, Codex, and ChatGPT, and full export and delete. Cancel anytime. It is a limited-time launch offer.

Yes. RaLHF is free for the first 90 days, no credit card required. After that it is $5 a month, and you can cancel anytime. The 90-day offer is a limited-time launch deal.

Create an account at app.ralhf.ai, connect a few apps or upload a document or two, then connect Claude, Codex, or ChatGPT. It takes a few minutes, and from then on the AI you use works with your context.

Yes. RaLHF Enterprise is a customer context platform for businesses building AI products. It can run in your own environment, connect to your CRM and other systems, and give your AI the customer context it needs to feel personal at scale. Contact info@botfood.ai for a demo.