Skip to content
CamouDatAI
HUEN
Live demo →

Features

What the platform can do

Eight functional areas, 69 capabilities presented. Capabilities marked NEW arrived with the v7 releases.

01

Chat & user experience

Secure AI conversation for everyday work, in Hungarian and English, on a clean interface.

Your day: upcoming meetings on the home screenNEW

Your own calendar’s next meetings appear on the home screen. You can join a Teams meeting with one click. A meeting that is about to start is highlighted. The calendar data never reaches the AI. So the day is clear at a glance.

Personal home screen: greeting, weather, name dayNEW

The home screen greets you personally: by time of day, holidays and name day, using your own form of address. The top bar also shows the weather and an hourly forecast. A daily tip, a monthly summary and pinned items wait there too. The text of the greeting never reaches the AI.

Personal memory: About me and Remember thisNEW

On the About me page you can set your form of address and a few lines about yourself, and the assistant takes them into account. At first sign-in a skippable step offers this too. You approve every Remember this proposal, and nothing is saved without you. The text is stored encrypted with your own key and reaches the AI masked.

Robust document handling across Office and PDF formatsNEW

You can convert a document in one request between the Office formats, PDF and CSV. A PDF can be translated with its original layout, and an editable Word file is produced alongside. On the whole sheet of a spreadsheet the assistant filters, groups and totals. Sensitive data goes out masked here too.

Code execution in an isolated sandbox

The assistant can also run a program to calculate or to produce a spreadsheet or document. The program runs in an isolated, network-free environment and cannot reach other clients or the system’s secrets. A conversation’s workspace persists within that conversation, expires later, and is deleted with the conversation. So the calculation is done by the program, not estimated by the model.

MS Office file editing and PDF creation

The assistant edits the Word, Excel and PowerPoint files you upload to the chat as you ask, and hands them back for download. A contract draft or a budget can be corrected without leaving the chat. The inserted text goes through a separate data-protection check. A chat answer can also be downloaded as a Word or PDF file.

Sharing a conversation within the organisationNEW

You can create a read-only link to your own conversation within the organisation. Whoever opens it sees the masked text, and only the messages written up to the moment the link was created. You can revoke the link at any time, and sharing is logged. So a good conversation can help colleagues too.

Managing capabilities (skills)

The company can install its own knowledge and working-method packages, called skills, for the assistant. On installation they go through the same check as uploaded files, and take effect only after approval. You can upload a skill of your own too. The Skill Creator turns a conversation into a skill, and the Document Co-author guides you through writing a document.

Secure AI conversation

Real-time, streaming answers on a clean interface, in Hungarian and English, with light and dark themes. Every request passes through the protection pipeline before anything leaves the organisation. Masking and content filtering run in the background, without slowing down the pace of the conversation. Users get the same convenience as a public AI tool, while company data stays inside.

Projects and workspaces

Conversations and their associated knowledge can be organised into projects, keeping different topics and clients separate. An existing conversation can also be moved into the right project from the sidebar with a simple drag and drop. Documents attached to a project are only available within that workspace. This means material from one client cannot leak into another client's conversation.

Automatic titles and tagging

The system automatically gives conversations a fitting title, so there is no need to name them by hand. The title is generated from the content, so the list still makes sense weeks later. Conversations can also be organised with tags, and tags can be used for filtering. When searching, title and tag narrow down the results together, which matters once there are many conversations.

Message editing and regeneration

A sent question can be edited afterwards and resubmitted, and the answer can be regenerated. A response that is still being generated can be stopped partway through if it is clearly heading in the wrong direction. Earlier versions are not lost, so the conversation can be reshaped freely. This means a longer task can be carried through in a single conversation, without starting a new thread.

Rich content and canvas

Answers appear with formatted text, code highlighting and mathematical formulas. In the canvas panel, longer texts can be edited in place, without needing to copy them into a separate editor. The finished material can be downloaded directly from there. The panel stays alongside the conversation, so the text and the question behind it are visible at the same time.

Conversation export to a branded document

A complete conversation can be downloaded as a tidy, company-branded PDF or Word document. The export preserves formatting, code snippets and source references. It is ideal for archiving, internal sharing or documenting a decision. The file can then be stored through the usual corporate channel, for example SharePoint.

Fast, responsive experience

Even long conversations scroll smoothly, and the first answer arrives quickly. The system caches repeated content, which reduces both waiting time and cost. The interface does not reload itself at every step, so work stays continuous. This is felt most clearly in daily, frequently repeated use.

Command palette

A single keyboard shortcut brings up a quick search for instant navigation. Conversations, menu items and common actions are all accessible from one place. Work can be carried out entirely with the keyboard, without needing the mouse. This benefits both heavy typists and accessible use alike.

Built-in bilingual help

There is no need to leave the interface: a side help panel explains what the given page does. It describes what can be done there and what is worth paying attention to. It is available across every area of the chat and the admin console, in Hungarian and English. This means onboarding does not depend on separate training material, since help is available right where the work happens.

Guidance on every control

Every switch and button has a short, plain-language explanation available. It can be brought up by touch or keyboard alike, so it also works on a phone. The text follows the language of the interface, in both Hungarian and English. This means no one has to guess what a setting does before changing it.

Installable web app

The interface also works as an installable web app (PWA), with its own icon on the device. It launches directly from the browser, without a separate installer package or store distribution. Localisation is complete, available in both Hungarian and English. The same protection rules apply here as in the browser interface.

02

Knowledge & documents

An assistant that works from your own documents and cites its sources.

Web access

When the company’s own knowledge is not enough, the assistant can run a public web search. By default AWS’s own search runs, and only the masked search phrase leaves. The administrator can also choose Brave, a third party outside AWS. With separate permission it also reads web pages and the National Legislation Repository. Every search is logged.

Answers from your own documents

The assistant finds its answers in your uploaded company documents, not on the internet. It marks which source it worked from, so the answer can be traced and verified. The search follows the user's own permissions, so it only surfaces what they would have access to anyway. This makes the answer accountable, not merely convincing.

Source preview in the answer

Hovering over a citation immediately shows a short excerpt of the source. There is no need to open the full document to verify a claim. The preview appears alongside the answer, so it does not interrupt reading. This speeds up verification most with long contracts and policies.

Broad document support

PDF, Word, Excel and PowerPoint files can be uploaded to the knowledge base, old Office formats included. The system extracts the text and screens it against the protection rules already at upload. Anything that does not comply never enters the searchable content. Material that is too large is processed in parts. So protection starts at the upload itself.

Scanned document recognition

The system recognises the text of scanned documents that arrive as images. Old paper-based material becomes searchable. The recognised text goes through the same protection check as other documents. Scanned PDFs in the knowledge base become searchable too.

Image analysis, only on verified clean images

Image files can be uploaded too: the system reads the text in them and passes it on masked. The image itself only goes out if it is verified to contain no sensitive data. In every other case only the masked text goes forward. Covered areas are checked again on the outgoing image.

"What goes out?": preview the outgoing image

With a single click, the user can see exactly the image the system would send to the model. The image is shown covered and stripped of metadata, that is, in the form it would actually leave in. This is a human checkpoint: automatic recognition does not always catch handwriting, faint text or faces. The user can see them, however, and can stop the send.

Deep research

For complex questions the system researches in several rounds. You choose among four sources, internal knowledge base, file server, SharePoint and OneDrive, and the public web. It gives a referenced answer backed by sources. It asks for confirmation only if the estimated cost is above the configured limit.

Reusable knowledge collections

Documents can be organised into thematic collections, with whole folders uploaded at once if needed. A single collection can be assigned to several conversations and projects. This way the assistant always works from the right knowledge base, not the entire archive. Maintenance also happens in one place, rather than separately for each conversation.

03

Enterprise data integrations

Connecting to existing enterprise systems, with each user's own permissions.

Microsoft 365, SharePoint, OneDrive

The assistant reaches SharePoint and OneDrive content with your own permissions, so it sees only what you are entitled to see. There is no separate service account with access to everything. A change in permissions applies from the next request. You can also attach a SharePoint or OneDrive file to the conversation, with the same checks as an upload.

Searching and reading Outlook mailNEW

The assistant can search your own mailbox and read messages, by sender, period or subject. It only reads: it does not send mail and cannot reach another mailbox. The text of messages reaches the AI masked. So you no longer have to dig out an old email by hand.

Calendar integration

Calendar reading happens directly from the conversation, through the Microsoft 365 account. Event handling is approval-gated, so the system never writes to the calendar on its own. Suggested time slots are generated from actual free and busy availability. This way scheduling can be handled in one place, without leaving the familiar calendar.

On-premise file server agent

An installable agent searches and reads on the company’s own file server, on the user’s behalf. It respects live permissions and gives no more than the colleague can already see. The index is encrypted, and the network does not have to be opened outward. The hit list can be exported to Excel or CSV, and a path can be copied in one click.

Semantic search on the file server too (RAG)

The file server agent also searches by content, not only by keyword. The search follows the user’s live permissions and brings a short text excerpt with each hit. The excerpt goes through the usual secret and data-protection filters, and the questions are logged. Files similar to a hit can be found, and more results can be requested.

SAP integration

Read-only SAP queries can be started directly from the conversation. Modifying operations need separate approval and never run on their own. Data-leak protection checks the result before anything reaches the AI. For SAP Business One there is also a separate read-only connector through the local agent.

Extensible connector catalogue

New enterprise systems can be connected to the platform through a uniform descriptor. Connectors can be enabled per tenant, so nothing switches on for everyone at once. Permission handling and logging apply to new sources in exactly the same way. This way the platform grows with the organisation's needs, without custom development.

04

Channels

CamouDatAI is available from the familiar Microsoft tools, with the same protection.

Microsoft Teams bot

The assistant can be used straight from Teams, with no separate interface. The protection and data-leak rules apply here too, and signing out also ends the session. In a personal chat you can search the company file server too. So staff do not have to get used to a new interface.

Proactive Teams notifications

The system can also send notifications in Teams on its own, not only respond to questions. For example, about a finished report or a completed workflow. The notification content goes through the same protection checks as the conversation. This way a colleague does not need to watch the interface to be informed in time.

Outlook add-in

AI assistance for e-mail directly in Outlook. Summarising, reply suggestions and extraction, without leaving the message. Sign-in uses the corporate account, with no separate password. The message content goes through masking just like any other request.

Office add-in

The assistant is built into the Word, Excel and PowerPoint interface. This makes it available while working on the document, without copying or switching windows. Selected content can be sent as a question, and the answer can be written back into the document. Protection checks run here too, before anything is sent.

Automatic user provisioning

Users and groups are created in sync from the corporate directory (SCIM). Departing colleagues' access is removed automatically through the same channel. There is no manual upkeep, so no orphaned accounts remain in the system. Group membership carries over to roles too, so permissions move together with the organisation.

05

Security & data protection

The heart of the product: protecting sensitive data at every point.

Claude on Amazon Bedrock, inside the EU

We use the Claude models through Amazon Bedrock, from an EU Region. Requests are processed only in AWS Regions inside the European Union. According to AWS, prompts and responses are not stored and are not used for model training. On top of that, only masked content ever reaches the model.

Automatic sensitive-data masking

The system detects personal and sensitive data and pseudonymises it before the request reaches the model. In the answer, the real values are automatically restored for the user. The AI provider therefore never sees the actual data, only the placeholder. The system also continuously and automatically checks that no internal path can bypass the masking layer.

Data loss prevention (DLP)

Customisable rules decide what may leave: block, warn or mask. The system checks responses even as they are generated, and stops them if it detects a leak. Rules can be set per tenant, so they can be tailored to the organisation's own risk profile. Every decision is logged, so afterwards it remains visible what was blocked and why.

Per-user encryption

Every user’s conversations are stored encrypted with their own key. No other user can look into them. Even the administrator can read in only with a reason, and it is logged. So protection does not depend on policy, but on how the data is stored.

Encrypted storage and key management

Every sensitive field is stored encrypted, and the master keys live in a hardware key store (HSM or Key Vault). Key rotation is regular and automatic, not a manual task. For client deployments, passwords and keys come from Azure Key Vault by default, not from the server's disk. This way a compromised server alone does not give up the secrets.

Sign-in protection: SSO, MFA, roles

Corporate single sign-on (Entra ID) with multi-factor authentication. Permissions can be set in a fine-grained, role-based system. The system strictly validates the identifiers received at sign-in, so nobody can get into another account with a similar but non-genuine identifier. There is no separate password on the platform, so there is no additional secret that can leak.

Confirmation for sensitive actions

Irreversible actions require a fresh, repeated corporate sign-in to start. This includes data deletion, the kill switch and compliance exports. Confirmation is required even when the user is already signed in. Sensitive actions sit on a single central list, and the system provably requires confirmation for every one of them, not just a few manually listed cases.

ITBS access in the customer's handsNEW

The customer’s administrator approves an ITBS operator access window, for a limited time. Approval needs a reason and a fresh confirmation, and the window closes by itself on expiry. The administrator can also close it earlier. Opening and closing are logged, and an open window is flagged in the admin interface.

Secret scanning in uploads

Passwords, keys and tokens accidentally left in uploaded files are filtered out automatically. A suspicious file goes no further, and the system raises an alert. The administrator can allow a false positive. The scan runs at the start of processing, so the secret never enters the searchable content.

Anomaly detection

The system watches for unusual usage patterns and flags anything suspicious. This can include a sudden spike in downloads or a bulk query starting at an unusual time. The goal is early incident detection, not after-the-fact explanation. The flag is also written to the log, so the investigation can be traced back.

Self-tuning protection proposals

The system generates proposals from the logs to refine protection rules. This can include adjusting thresholds or reducing false alarms. Rollout is always subject to human approval; nothing takes effect automatically. Accepting a proposal that weakens protection requires a separate, explicit confirmation, together with the displayed warning text.

06

Compliance & governance

Transparency, accountability and cost control, in the language of management and auditors.

Tamper-evident audit log

Every significant event enters an immutable, hash-chained log, sealed daily. This makes it possible to prove afterwards that the log has not been tampered with. A break in the chain is visible immediately, with no separate investigation required. On request, an independent witness can also be included alongside the daily seal.

Compliance pack in one click

A NIS2- and ISO-oriented compliance report and evidence pack export in a single operation. The pack includes the evidence itself, not just the claims. So preparing for an audit is not weeks of data gathering. The export itself counts as a sensitive operation, so it requires a fresh sign-in.

Retention rules and proof of deletion

Data retention periods are configurable, and deletion is verified with proof. This directly serves GDPR expectations. Expired entries in the audit logs are genuinely removed by a scheduled process. This is not a documented intention but an operation that actually runs on every deployment.

Policy acceptance is provable

Acceptance of the usage policy leaves a server-side trail that cannot be forged after the fact. It records who accepted which version, and when. Per-user history can be looked up in the admin interface. So a disputed case does not have to rely on memory.

Scheduled reports by e-mail

Scheduled management and operational reports are generated at set times. The report lands in the responsible person's mailbox, prepared and ready for review. Sending remains a human step: the system does not send e-mail out on its own. So the content can be reviewed once more before it reaches anyone.

Cost forecasting and quotas

The system forecasts the expected AI cost from measured consumption. Limits and alerts can be set for groups and for individual users, even a custom monthly limit. Users can see their own consumption too. So there is no surprise on the bill at the end of the month.

Admin console and kill switch

A full management interface covers settings, the dictionary, rules and users. Every change is logged, so it remains visible afterwards who changed what. If needed, outbound AI traffic can be stopped with a single switch. The kill switch requires a fresh sign-in so it cannot be triggered by accident.

Live inspector and demo mode

Admins can follow in real time how a request moves through the protection pipeline. It shows what the system recognised, what it masked and what it let through. This is also excellent for demonstrations and internal training. The inspector shows real behaviour, not a pre-recorded illustration.

07

Automation & workflows

Automating repetitive work, always inside the protection boundary.

Workflow templates

Multi-step, recurring tasks are organised into declarative chains. With management approval, a chain can continue running automatically, so not every step needs to be started by hand. Templates are reusable, so a proven process can be handed on to colleagues. Every step goes through the same protection checks as a manually started request.

Flow editor with decision branches

Automation is built by clicking, with no coding. The flow can also decide, and the two branches are shown in separate lanes. Web search, change watching and e-mail to your own mailbox can be added as steps. The editor only offers what the system can carry out, and flags errors before saving.

Scheduled runs, requested and approved

Flows can also run on a schedule: daily, weekly or monthly. Staff request the schedule, and an administrator approves it. Only someone whose request was approved for it can send out an email or a document. If a scheduled run is missed, the interface warns you.

Output actions

The finished result can be saved to SharePoint, posted to Teams, or written into an Outlook draft with a single button. Outgoing content goes through the same protection checks here too; it does not bypass the rules. A draft stays a draft: actually sending it is a human step. So automation prepares, but does not decide on behalf of the organisation.

Meeting notes

The system turns transcripts of Teams meetings into structured notes. Action items are collected into a separate list, with an owner and a deadline where these were mentioned. The notes can be exported or shared in the usual way. Processing the transcript goes through the same masking as any other content.

08

Platform & reliability

Enterprise-grade foundations working in the background, for trust and continuity.

Custom branding

The interface can be tailored to the client's branding: its own name, logo and colours. This lets CamouDatAI appear to staff as the partner's own solution. The branding also carries through to exported documents and to reports that are sent out. The underlying operation and protection rules stay unchanged.

Flexible model choice

The system works with several AI providers, for example Anthropic models or a European Bedrock route. The choice can be matched to data-residency needs. Changing the model does not affect masking and logging. If the requested model is temporarily unavailable, another permitted model answers, and the interface says so.

Reliable operations

Automatic backup and restore, together with continuous monitoring, look after continuity. Outbound network traffic is strictly restricted, and only approved destinations are reachable. The server configuration is hardened, and unnecessary services are switched off. Operation is measured, so faults do not surface only through user complaints.

Capacity guard for heavy processing

Image recognition is resource-intensive, so the system limits how much heavy work can run at once. This means a single large upload does not slow down other users. Waiting jobs are queued rather than lost. The cap never lets data out unchecked: if there is no capacity, the request waits, rather than proceeding without review.

Infrastructure as code

The whole environment can be deployed from declarative, versioned code. A new instance can thus be built predictably, with the same configuration. Changes can be reviewed and rolled back like any other code. This is what makes deployment repeatable and auditable.