What is mAItflow? Features, AI agents and use cases
mAItflow is an AI workspace for companies, developed and operated by Masterplan Tech Solutions GmbH in Munich, Germany. The browser-based platform connects an organisation's knowledge – documents, meetings and connected business systems – with specialised AI agents. Teams use it to find answers with source references, create documents, presentations and analyses, follow up on meetings and organise projects. Results stay reviewable: content can be edited before it is sent, and actions in external systems require confirmation.
In brief
- Category: an AI workspace for companies – a browser-based cloud application with specialised AI agents instead of a single chatbot.
- Core features: questions to company knowledge with source references, documents (DOCX), presentations (PPTX), spreadsheets and analyses, meeting follow-up, projects, research and marketing.
- Audience: according to the provider, mainly small and mid-sized companies and teams that want to use their knowledge and automate recurring tasks.
- Typical workflow: a recorded meeting becomes minutes, a task list, a project and a presentation.
- Control and data protection: operated in the EU, results reviewable before they are sent, actions in external systems only after confirmation; data processing agreement (DPA) and TOMs on request.
- Limitations: AI results must be reviewed, the available features depend on plan and configuration, and answers are only as good as the connected data.
mAItflow at a glance
| Attribute | Details |
|---|---|
| Product name | mAItflow |
| Category | AI workspace for companies (agentic AI workspace) |
| Provider and operator | Masterplan Tech Solutions GmbH, Leopoldstr. 31, 80802 Munich, Germany; Munich Local Court, HRB 307911 (legal notice) |
| Audience | Small and mid-sized companies and teams (as stated by the provider) |
| Main capabilities | AI Chat with source references, AI agents, knowledge base, documents, presentations, spreadsheets and analyses, meetings, projects, research, marketing, value tracking (ROI) |
| AI agents (selection) | Sage, Sven, Silas, Lexi, Artemis, Atlas, Nico, Rico, Mara (agent profiles) |
| Company knowledge | Uploaded files, meeting transcripts and data synced through integrations |
| AI models | Models from third-party providers: Azure OpenAI, OpenAI, Anthropic Claude, Google Gemini, Mistral (AI transparency) |
| Languages | User interface and website in German, English and French |
| Deployment | Browser-based cloud application (software as a service) |
| Data processing | Operated in the EU; depending on the AI model provider selected, processing outside the EU is possible |
| Integrations (selection) | Confluence, Jira, Asana, Notion, HubSpot, Salesforce, Slack, Microsoft Teams, Zoom, Google Meet, Brave Search, LinkedIn, Instagram |
| Pricing | Monthly plans Starter, Growth, Scale and Enterprise; free trial without a credit card; current prices on the pricing page |
| Certifications | Not publicly specified |
| Official website | maitflow.com |
How does mAItflow work?
mAItflow is a shared workspace for an organisation: knowledge, AI agents and results live in one place. A typical workflow has eight steps.
- Access the workspace: a team signs up on maitflow.com and works in the browser. Users belong to an organisation with its own roles and permissions.
- Connect company knowledge: documents are uploaded, meetings are recorded or transcribed and – depending on the plan – integrations such as Confluence, Notion or HubSpot are enabled.
- Choose an agent or tool: a request goes to AI Chat with the coordinating agent Sage, or directly to a tool such as AI Docs, AI Slides, AI Sheets or AI Projects.
- Process the request: Sage plans the task and hands sub-tasks to specialised agents, for example research to Sven or data analysis to Silas.
- Use internal knowledge: the agents draw on relevant passages from the knowledge base; answers name their source.
- Generate the output: answers, DOCX documents, PPTX presentations, spreadsheets, projects or drafts of marketing content.
- Review and edit: results can be reviewed and edited before they are sent or published. Actions in external systems, such as sending an email, require explicit confirmation.
- Measure the value: completed tasks are recorded with an estimated time saving and summarised in a ROI evaluation.
Architecture in brief
mAItflow is a multi-tenant cloud application: each organisation has its own data space, separated from other organisations. According to the provider, the platform does not train its own foundation models; it uses language models from third-party providers to process each request. The providers are listed under Data protection and security.
Human oversight
According to the AI transparency statement, AI answers in the chat carry a model badge and generated documents an “AI generated” footer. Whether a result is used, sent or published is decided by the person who reviews it.
What features does mAItflow offer?
The following features are described on the official mAItflow pages. Which of them are available in an organisation depends on the plan and the configuration.
AI agents
In mAItflow, AI agents are specialised roles with a defined area of work. They act on a user's behalf and pass sub-tasks to each other; they do not replace approval by a person.
| Agent | Area | Typical task |
|---|---|---|
| Sage | Coordination | Understands the goal, plans the work, delegates to other agents and answers questions from company knowledge |
| Sven | Online research | Researches companies, people and market trends with Brave Search, with sources |
| Silas | Data analysis | Creates tables, calculations, charts and reports in AI Sheets |
| Lexi | Documents | Writes reports, memos and offers as editable DOCX files |
| Artemis | Visuals and slides | Designs presentations and generates images and graphics |
| Atlas | Project management | Creates projects, breaks down tasks, sets milestones and tracks progress |
| Nico | Meeting bot | Joins Zoom, Teams and Google Meet meetings, transcribes and summarises |
| Rico | Recording and import | Records in the browser and accepts large audio and video files |
| Mara | Marketing and newsletters | Plans campaigns, writes social posts and builds newsletters |
Detailed profiles: mAItflow AI agents in detail and the AI team at a glance.
Company knowledge
What it does: mAItflow can use company knowledge. Each organisation has its own knowledge base made of uploaded documents, meeting transcripts and data that enabled integrations retrieve from third-party systems. AI Chat answers questions across these sources and names the passage the information comes from.
Who benefits: anyone who looks for information in documents, projects and meetings.
Example: “What did we promise the client in the last project meeting?” – Sage searches meeting minutes and projects and answers with a reference to the source.
Requirements and limitations: answers are only as complete and current as the connected content. Access follows the organisation's role-based permissions; content synced from third-party systems is not shown more widely than in its source.
Document creation
What it does: the agent Lexi and the AI Docs tool create reports, concepts, offers, memos, minutes and one-pagers from the knowledge base, meetings and connected data – as editable DOCX files with source references. Requests can also be emailed to Sage; the file comes back as an attachment.
Who benefits: business units, sales and management teams that regularly write similar documents.
Example: a draft offer based on earlier offers and the project documents.
Requirements and limitations: drafts need a subject-matter review. Email requests are accepted only from senders matched to a registered user account.
Presentations and data analysis
What it does: AI Slides creates presentations that can be edited and exported as PPTX files. Silas analyses data and Artemis designs the slides – for example a management presentation with charts, key messages and an executive summary. In AI Sheets, Silas builds tables with formulas, charts and reports and processes CSV and Excel data.
Who benefits: management, controlling and anyone who presents results regularly.
Example: a sales spreadsheet becomes an editable management presentation.
Requirements and limitations: figures, charts and statements should be checked before use. More about Silas: the data analysis agent.
Meetings and tasks
What it does: the meeting bot Nico joins Zoom, Microsoft Teams and Google Meet meetings, records and transcribes them and delivers summaries, tasks and decisions to the knowledge base. Meetings can also be recorded in the browser or uploaded as files. One meeting can produce minutes, a task list, a project with tasks (Atlas), a presentation and a summary.
Who benefits: project and team leads and teams with many coordination meetings.
Example: after a kick-off meeting, the minutes, a task list and a new project are ready.
Requirements and limitations: recordings require the participants' consent. Transcripts and summaries should be checked for accuracy. Availability of the meeting bot depends on the plan.
Research and marketing
What it does: Sven researches the web with Brave Search and returns results with sources. Mara plans marketing campaigns, writes social posts, generates images, tracks trends and competitors and builds newsletters for CRM recipient lists.
Who benefits: sales, marketing and business development.
Requirements and limitations: research results depend on the sources found and need checking. Social posts and newsletters can be reviewed and edited before publication. More: Sven and Mara.
Workflow automation
What can be automated: multi-step work distributed across agents – for example meeting, minutes, tasks, project and presentation – and recurring processes through the automation-workflow tool that the pricing overview lists depending on the plan. Requests emailed to Sage are part of this as well.
What does not run fully automatically: actions in external systems, such as sending an email or changing a calendar entry, require explicit confirmation. More on the principle: human-in-the-loop for agentic AI.
Productivity and ROI measurement
What it does: every task an agent completes is recorded with an estimated time saving. The evaluation shows hours saved, costs saved, usage per team, the most-used agents and the trend over time.
How it is calculated: estimated value = estimated effort in minutes ÷ 60 × hourly rate, minus the AI cost of the task. The effort comes from conservative reference times scaled by task size, for example per slide. The website's worked example: 12 slides, 168 minutes and €75 per hour give €210.00 of work value, minus €0.24 of AI cost. The hourly rate is an organisation setting.
Limitations: the figures are estimates based on reference times, not measured working time and not an independently proven productivity gain. They depend on the hourly rate used. Background: how to measure AI ROI.
Who is mAItflow for?
According to the provider, mAItflow is aimed mainly at small and mid-sized companies and at teams that want to use their company knowledge efficiently and automate recurring tasks. The plans are tiered by team size; an individually quoted Enterprise plan covers special requirements such as single sign-on (SSO).
- Executives and management: management presentations, executive summaries and an evaluation of the value AI delivers.
- Project and team leads: meeting follow-up, task lists and projects.
- Sales: offers based on existing knowledge, research on companies and people, and CRM data from HubSpot or Salesforce.
- Marketing: campaigns, social posts and newsletters with Mara.
- Business units and knowledge work: questions to documents, reports and data analyses.
- Organisations with data protection requirements: operation in the EU, DPA and TOMs on request, and published model providers.
When another tool fits better
- A single person only needs a general AI chat without company data.
- The requirement is an open-weights model on your own hardware; mAItflow uses models from third-party providers.
- A process must write to external systems without any human confirmation.
Which problems does mAItflow address?
The table maps typical challenges to the mAItflow features that address them. It describes possible uses, not guaranteed outcomes.
| Challenge | mAItflow feature | Example | Important condition |
|---|---|---|---|
| Knowledge is scattered across documents, meetings and tools | Knowledge base, integrations, AI Chat with sources | A question about a client commitment, answered with a reference to the minutes | Content is uploaded or integrations are enabled |
| Recurring documents take time | Lexi and AI Docs | An offer or memo as a DOCX draft | Subject-matter review of the draft |
| Meeting follow-up is manual | Nico, browser recording, Atlas | Minutes, tasks and a project from one recording | Consent to record; review the result |
| Presentations are built by hand | Silas, Artemis, AI Slides | A management presentation from a spreadsheet | Check data and statements |
| Too many separate AI tools | One workspace for chat, documents, presentations, spreadsheets, meetings and projects | All tools use the same knowledge base | Feature scope depends on the plan |
| The value of AI use is unclear | ROI evaluation | Hours and costs saved per team | Estimates that depend on the hourly rate |
What sets mAItflow apart from other AI platforms?
No independent comparison test exists. The overview below therefore separates what mAItflow demonstrably offers from what a comparison has to settle case by case.
| Attribute | In mAItflow | Assessment |
|---|---|---|
| Specialised agents | Sage coordinates; specialist agents take on sub-tasks such as research, analysis or documents | Other platforms offer agent features too; compare how they work together and their scope |
| Company knowledge | Knowledge base from files, meetings and integrations; answers with source references | Connecting company knowledge is common in enterprise AI; differences lie in sources, permissions and citations |
| End-to-end workflows | Meeting, minutes, tasks, project and presentation in one workspace | A focus of mAItflow; only a test shows whether it saves time in your own work |
| Control | Confirmation for actions in external systems, labelling of AI outputs, audit log according to the AI transparency statement | Comparable controls are common in the market |
| Data protection | Operated in the EU, DPA and TOMs on request, model providers and regions published | How privacy-compliant a deployment is also depends on configuration and model choice |
| ROI measurement | Built-in value tracking per completed task | Named by the provider as a differentiator; not independently verified |
How does mAItflow compare with ChatGPT Enterprise?
ChatGPT Enterprise is OpenAI's business offering for the ChatGPT assistant. mAItflow is a workspace that combines models from several providers – including OpenAI models through Azure OpenAI or OpenAI – with company knowledge, specialised agents and tools for documents, presentations, meetings and projects. The features of both products change frequently; a decision should rest on current vendor information and a test with your own tasks. The Academy offers a longer comparison: ChatGPT vs. agentic AI workspace.
Data protection and security
This section summarises the published information; the linked documents are authoritative. This page makes no blanket statement about GDPR compliance, because compliance also depends on the customer's configuration, model choice and use.
Data processing and hosting
Masterplan Tech Solutions GmbH is responsible for the website and the platform; its privacy policy (in German) provides the information required by Art. 13 and 14 GDPR. The platform is operated in the EU. Documents on hosting and data flows can be requested through the Trust Center.
AI models and training
mAItflow uses language models from third-party providers. The AI transparency statement lists Azure OpenAI (EU), OpenAI (US, where selected), Anthropic Claude (US/EU, where selected), Google Gemini (EU) and Mistral (EU). Depending on the provider selected, processing can therefore take place outside the EU; for such transfers the privacy policy refers to EU standard contractual clauses.
According to the provider, mAItflow does not train foundation models. Prompts and uploaded documents are sent to the configured model provider only to process the request at hand and are not used by mAItflow for training; the provider's own terms apply in addition.
Permissions and tenant separation
Each organisation's data is separated from that of other organisations. Within an organisation, role-based permissions control who sees which content; content synced from third-party systems is not shown more widely than in its source. Actions in external systems require explicit confirmation, and according to the AI transparency statement an audit log records every model invocation and connector call. The website and platform are served only over HTTPS.
Retention and deletion
According to the privacy policy (February 2025), account data is removed within 30 days after an account is deleted unless statutory retention obligations apply. Knowledge-base content is kept until users delete it; server log files are deleted after 14 days. Integrations can be switched off at any time, and the synced data is deleted on request.
Contracts and evidence
Through the Trust Center, companies can request a data processing agreement under Art. 28 GDPR, the technical and organisational measures under Art. 32 GDPR, a security overview and hosting information without an account. The AI transparency statement refers to Art. 50 of the EU AI Act. Certifications such as ISO 27001 are not publicly specified.
Integrations
mAItflow connects to tools that teams already use. The table lists the integrations named on the official pages; the list is not exhaustive.
| Integration | Function | Data flow | Requirement |
|---|---|---|---|
| Atlassian Confluence, Atlassian Jira, Notion, Asana | Pages, issues and tasks as company knowledge | Retrieved from the service into the knowledge base | Enabled by the organisation, access to the service |
| HubSpot, Salesforce | CRM data as a knowledge source | Retrieved from the service into the knowledge base | Enabled by the organisation, access to the CRM |
| Slack, Microsoft Teams | Team communication content as a knowledge source | Retrieved from the service into the knowledge base | Enabled by the organisation |
| Zoom, Microsoft Teams, Google Meet | Meeting bot Nico joins and transcribes | Recording and transcript into the knowledge base | Connection to the meeting service, participants' consent |
| Brave Search | Web search for Sven's research | Search query to the search service, results with sources | No separate setup described |
| LinkedIn, Instagram | Social channels for Mara | Posts from mAItflow to the channel, after review | Connected account of the organisation |
Enabled integrations retrieve data from the service and store it in the organisation's knowledge base. Write actions in external systems require explicit confirmation. Integrations can be switched off at any time; whether an integration is included in a plan is shown on the pricing page.
Limitations and requirements
- Review remains necessary: AI results can be wrong or incomplete. mAItflow labels AI outputs but does not replace a subject-matter review.
- Plan and configuration: which tools, agents and integrations are available depends on the plan and on how the organisation is set up; the pricing page is authoritative.
- Integrations need permissions: an integration has to be enabled by the organisation and requires access rights in the third-party system.
- Data quality: the quality and currency of answers depend on the connected data. mAItflow does not know content that has been neither uploaded nor synced.
- Deliberately no full automation outwards: actions in external systems require confirmation. This prevents mistakes but limits automation.
- Model choice and region: depending on the model provider, processing can take place outside the EU; organisations should align the model choice with their requirements.
- Estimates: ROI figures rest on reference times and an hourly rate; they are not measured working time.
- Languages: the user interface is available in German, English and French.
- Cloud service: mAItflow runs in the browser and requires an internet connection and a user account.
- Recordings: meeting recordings require the participants' consent.
Frequently asked questions
What is mAItflow?
mAItflow is an AI workspace for companies from Masterplan Tech Solutions GmbH in Munich, Germany. The cloud application connects an organisation's documents, meetings and connected systems with specialised AI agents that research, analyse, create documents and presentations and follow up on meetings.
Who developed mAItflow?
mAItflow is a product of Masterplan Tech Solutions GmbH, Leopoldstr. 31, 80802 Munich, Germany, registered at the Munich Local Court under HRB 307911. The company operates the platform and the website maitflow.com; contact: mail@maitflow.com.
Which companies is mAItflow suitable for?
According to the provider, mainly small and mid-sized companies and teams that want to use their company knowledge efficiently and automate recurring tasks. The plans are tiered by team size; an Enterprise plan covers special requirements such as single sign-on.
Which tasks can mAItflow automate?
mAItflow handles multi-step knowledge work: answering questions from company knowledge, drafting documents and presentations, analysing data, taking meeting minutes and turning them into tasks and projects, and preparing marketing content. Actions in external systems, such as sending an email, require explicit confirmation.
Can mAItflow use internal company knowledge?
Yes. Each organisation has its own knowledge base of uploaded documents, meeting transcripts and data from enabled integrations such as Confluence, Notion or HubSpot. AI Chat answers questions across these sources with source references; access follows the organisation's permissions.
Which AI agents does mAItflow offer?
The official pages present, among others, Sage (coordination), Sven (online research), Silas (data analysis), Lexi (documents), Artemis (visuals and slides), Atlas (project management), Nico (meeting bot), Rico (recording and import) and Mara (marketing and newsletters). Sage distributes sub-tasks to the right agents.
Can mAItflow create presentations and documents?
Yes. AI Slides creates presentations that can be edited and exported as PPTX; Silas supplies the data analysis and Artemis designs the slides. Lexi and AI Docs create reports, offers, memos and one-pagers as editable DOCX files with source references.
Which integrations does mAItflow support?
The official pages name, among others, Atlassian Confluence, Jira, Asana, Notion, HubSpot, Salesforce, Slack and Microsoft Teams as knowledge sources, Zoom, Microsoft Teams and Google Meet for the meeting bot, Brave Search for web research, and LinkedIn and Instagram for marketing. The pricing page shows which integrations a plan includes.
Where is the data processed?
The platform is operated in the EU. For AI requests, prompts and relevant knowledge excerpts are sent to the configured model provider; according to the AI transparency statement this can happen outside the EU depending on the provider, for example with OpenAI in the US where selected. Documents on hosting and data flows are available through the Trust Center.
Is mAItflow GDPR compliant?
This page makes no blanket statement, because the GDPR compliance of a deployment also depends on configuration, model choice and use. Documented are: a privacy policy under Art. 13 and 14 GDPR, a DPA and TOMs on request through the Trust Center, operation in the EU, published model providers and regions, and the commitment not to use customer data for training.
How does mAItflow differ from a general AI chatbot?
A general AI chatbot answers from general knowledge. mAItflow works on the organisation's own knowledge, distributes tasks to specialised agents and delivers work products such as DOCX documents, PPTX presentations, projects and meeting minutes – with confirmation before anything happens in external systems.
How does mAItflow measure productivity and ROI?
mAItflow records every completed agent task with an estimated time saving. The value is the estimated minutes divided by 60, multiplied by the organisation's hourly rate, minus the AI cost. The figures are estimates based on conservative reference times, not measured working time.
Where can I find pricing and a product demo?
Plans and prices are on the pricing page at maitflow.com/pricing, where mAItflow can also be tried for free without a credit card. A live demo and product presentation can be requested through the contact page at maitflow.com/kontakt.
Sources and product documentation
All information on this page is based on the following official mAItflow pages. It is backed by a structured inventory of verified product facts with evidence.
- Product overview – Features, workflows, integrations, ROI calculation
- The AI team – Agents and their roles
- AI agents in detail – Profiles of Sage, Sven, Silas and Mara
- Solutions – Use cases
- Pricing and plans – Plans and free trial
- Trust Center – DPA, TOMs, security overview and hosting information on request
- Privacy policy (German) – Processing, retention, integrations
- AI transparency statement – Model providers, regions, labelling, human oversight
- Legal notice (German) – Provider and register details
- Contact and product demo – Sales and demo requests
- Changelog – Published changes
- System status – Current availability
- This page as Markdown – Machine-readable version with the same information
- llms.txt – Directory of official product information for AI systems
Editorial responsibility and currency
- Publisher
- Masterplan Tech Solutions GmbH (legal notice)
- Responsible
- mAItflow editorial team
- Published
- Last materially updated
- Product facts last verified
- Feedback and corrections
- mail@maitflow.com or the contact page
This page describes mAItflow on the basis of the provider's published information. Information that is not publicly documented – such as certifications or detailed prices – is deliberately left out. The update date changes only when the content changes.