Help & features.

Getting started

1. Sign in with your email. If your organization’s domain is registered for Microsoft SSO, Storm2Flow redirects you to your tenant’s Microsoft login automatically; otherwise the password field appears so you can sign in with an email/password account.
2. Create a process by clicking the + button in the process library.
3. Describe your process in the text editor using plain language.
4. Select diagram types (BPMN, flowchart, swimlane, sequence, mind map).
5. Click Generate to create your diagrams.

That’s it. You can then edit, export, share, and version your diagrams.

Connect Microsoft sign-in for your company. If your organization is eligible, you may see an option to set up Microsoft (Entra / Azure AD) sign-in for your whole team. You start the setup, then forward the generated approval link to your company’s Microsoft admin. Once they approve, your colleagues can sign in with Microsoft and you become the organization admin. Nothing changes until your Microsoft admin approves.

Workspace & bookmarks

The editor workspace is built from collapsible panels, each shown as an edge bookmark so you can shape your own focus around the work you are doing.

Right-edge bookmarks

The right edge holds three inspector panels as bookmarks - Comments, Diagrams, and Assets - and only one is open at a time: opening one collapses the others. The Diagrams bookmark is the diagram-type switcher (BPMN, flowchart, swimlane, sequence, mind map). Storm2Flow remembers which panel you had open per process.

Left-edge bookmarks

The left edge holds two panels that each collapse to a bookmark: All Processes (your process library) and the current process description, labelled with the process name so the bookmark tells you which document it reopens. Both can be open at the same time; collapsing both frees the canvas.

Diagram fullscreen

The fullscreen control (⛶) in the diagram toolbar collapses both left panels so the diagram fills the frame. Toggle it off again to restore your previous layout.

Description editor

Write your process description in the rich text editor. A small toolbar above the editor gives you bold, italic, heading, bullet list, and numbered list buttons; the editor saves the underlying markdown so what you type round-trips losslessly through Storm2Flow.

The action row below the editor has three buttons: Analyse (described below), Save (commit your current text as a restorable version, a no-op for history if nothing changed, but always confirms with a toast), and Generate.

Analyse

Click Analyse to have the AI review your description for grammar, completeness, and clarity. If gaps are found, a Q&A panel appears with clarifying questions. Answer them and click Apply answers to improve the description.

Voice input

Click the Record button to dictate your process. A recording modal opens with a live timer and two actions: Discard to throw the take away, or Stop & transcribe to use it. While you are recording, Analyse and Generate are blocked so you cannot act on a half-captured description by mistake. After you stop, a spinner shows the take being transcribed; the transcript stays in the language you spoke (auto-detected: English, German, Portuguese, Spanish, French) and is routed through the description review below. Requires microphone permission.

Image & file upload

Upload a photo of a whiteboard, Miro board, hand-drawn flowchart, PDF, or .docx / .xlsx file. The AI extracts the process from the source and routes the extracted text through the content integration review below.

Braindump

Click Braindump next to Upload file to paste raw notes, meeting minutes, or brainstorm output for the current process. The text is routed through the content integration review below; the raw braindump is not persisted on its own - only the accepted enriched description is saved.

Description review

Voice, file/image upload, braindump, and structural edits to a BPMN diagram all converge on the same review modal - one consistent “review the suggested description before it lands” experience:

1. Your current description is saved as a draft.
2. The new content (or the description Storm2Flow derived from your diagram edit) plus your current description go to the AI together.
3. The review modal shows the updated description + a short change summary (1–7 bullets) and a diff against the previous text. Switch between Clean view and With highlights view to see exactly what changed.
4. Accept saves the suggestion as a new description version; Skip discards it. You can also edit the suggestion before accepting right in the modal - the edit field is a formatted editor with a small toolbar (bold, italic, heading, bullet list, numbered list), so you tweak the text in place without seeing any markup source.

Braindump and transcription results reveal progressively as the AI works, instead of popping in fully formed, so you can see the system is in motion. If the editor is empty when you add new content, the content becomes the first draft directly, and you still see the same review modal so you can back out.

Description versioning

Every saved change to a description becomes a restorable version - nothing is silently overwritten. Whether the change comes from accepting a review suggestion, editing the description directly, or generating diagrams, the prior text is preserved in version history and can be restored at any time.

Comments

Open the Comments bookmark on the right edge to discuss a process with your team. A comment can be anchored to a span of the description (select text first, then comment, and the comment quotes that span), pinned to a step (click a node on the diagram, then comment), or made about the whole process. Any comment can carry an optional Improvement tag, which marks it as an improvement idea for the process. A filter at the top of the panel switches between All, Open, Resolved, and Improvements, and comments can be resolved and reopened. The old standalone “Improve this process” notes box is retired - its job is now a whole-process comment with the Improvement tag.

Flagging a comment

Every comment row has an inline flag icon that sets a colour on the comment. Red is the one colour with a fixed meaning: red is a “Red flag”, a risk, gap, or concern you want to call out. The other colours carry no built-in meaning and are yours to use for grouping or styling threads however you like.

Comments and flags on the diagram

When you click Generate again, your open comments are placed directly on the diagram as comment annotations attached to the step they are about, never as extra boxes in the flow. Each one renders as the notation’s native comment primitive: a note or brace on Mermaid flowcharts and swimlanes, and a Text Annotation joined by a dashed line on BPMN. So a comment reads as a note beside the flow, not as another step. A flagged comment carries its colour onto the diagram where the notation allows it, so a Red flag still reads as a concern.

Where each comment lands:

• A comment on a step (created by clicking a node) lands on that step.
• A comment on a highlighted passage of the description lands on the step the passage maps to. Storm2Flow matches the words first, then falls back to AI for passages that do not literally repeat a step’s words; a passage with no confident match is simply left off the diagram.
• A comment on the whole process lands at the end of the flow.

Comments only reach the diagram on a Generate you trigger, the same way an edit to your description does. Adding or flagging a comment never changes your current diagram on its own, and there are no extra AI calls: the open comments simply ride along with your next Generate. This works on flowcharts, swimlanes, and BPMN.

Add subprocess

For any leaf process (one with no children), the editor header shows ＋ Add subprocess next to ⎇ Branch into As-is / To-be. Click it to decompose the current process into a more detailed sub-step. The subprocess hangs under the current one in your sidebar; navigating to it opens a fresh editor where you can describe and generate diagrams independently. Subprocesses are allowed inside As-is / To-be branches.

Try a sample process

First-time users see a starter gallery in the empty editor. Pick a sample (Leave Request, Customer Onboarding, Purchase Order, Customer Support, Software Architecture, Insurance Claim, Initiative Cockpit) and Storm2Flow creates a real process with the sample description and runs the diagram generation. You land in a working editor instead of a blank canvas. The gallery hides as soon as you have your own processes.

The same catalog is always reachable from the Examples button in the header - open it any time to spin up another sample without clearing your existing processes.

About AWS Bedrock

Storm2Flow runs every AI call on AWS Bedrock, hosted in eu-central-1 (Frankfurt) with cross-region inference profiles restricted to EU regions only. No customer description, image upload, voice transcript, file extract, braindump, or generated diagram is sent to OpenAI’s SaaS endpoints, Anthropic’s SaaS endpoints, or any non-EU model host. This is what makes Storm2Flow usable by EU organisations under DSGVO/GDPR without a separate data-processing addendum per provider.

Bedrock-only as of 2026-05-18. Earlier Storm2Flow versions had the diagram cascade on Bedrock but a few side surfaces (analyse description, integrate voice / file / image / braindump content, describe-from-XML, vision extraction, BPMN edit/delta) still called Anthropic and OpenAI directly. The PLAN-BEDROCK-ONLY migration moved every last surface to Bedrock; the @anthropic-ai/sdk and openai packages are no longer in the backend dependency tree. There is no longer any code path in the product that can reach api.anthropic.com or api.openai.com.

Why Bedrock matters to you

• EU data residency by default. Every model call - from the description-to-graph extraction to the layout pass - happens inside an EU AWS region. No transatlantic data egress, no per-provider DPA to negotiate.
• Cascade fallback - resilience without setup. Behind the scenes, fresh-diagram generation runs through a multi-rung Bedrock cascade. If the first model can’t produce a coherent diagram from your description, the next one tries, then the next. From your perspective the cascade is invisible - just better odds of getting a diagram on the first generate. When the second or third rung wins, the admin Diag panel surfaces it as winningRung: 2 or 3.
• One configured rank, no per-customer keys. You don’t paste an API key, you don’t pick a model, you don’t worry about provider quotas. The cascade is configured server-side and the same rank applies to every customer in the same Storm2Flow tenant. Less to set up; one audit boundary in CloudTrail.

For organisations that need every Bedrock call to land on their own AWS bill / their own CloudTrail (BYO Bedrock via cross-account role assumption), see Administration below - this is on the post-spike roadmap. The internal call path already accepts a per-request injected BedrockRuntimeClient, so the customer-facing UI is the remaining work.

Viewing & navigation

Zoom

• Ctrl + scroll wheel (or Cmd + scroll on Mac) to zoom in/out.
• Use the + / − / Reset buttons in the toolbar.

Pan

• BPMN: click and drag the canvas background to pan.
• Arrow keys: ← → ↑ ↓ pan by 50px.
• Mermaid diagrams: scroll normally to pan.

XML / code view

Click XML (for BPMN) or Code (for Mermaid types) to see the raw diagram code. Click Copy to copy it to your clipboard.

Tabs

When multiple diagram types are generated, tabs appear at the top of the diagram area. Click a tab to switch between types.

Editing Mermaid diagrams - flowchart, swimlane, mind map

Flowchart, swimlane, and mind map diagrams can be edited visually in the browser:

1. Click Edit to open the visual editor.
2. Drag nodes to reposition them. Connections are automatically preserved.
3. Double-click a node to rename it inline.
4. Use the toolbar to Add step, Add decision, Add circle, or Delete selected nodes.
5. Click Save to apply changes as a new version, or Cancel to discard.

If your edits change the process structure, the AI may suggest an updated description to keep text and diagram in sync.

Rendering: flowchart, swimlane and mind map now use the same ReactFlow + ELK renderer for view and edit mode, so the canvas looks identical whether you are just viewing or actively editing. Sequence diagrams continue to use the classic Mermaid renderer - their visual grammar doesn’t map cleanly to a graph canvas.

For swimlanes, Storm2Flow saves lane identifiers in canonical Mermaid form behind the scenes while preserving readable lane labels, so edit, save, reload, and re-edit cycles keep lane containers intact.

Connecting and re-routing edges (draw.io-style)

Every flowchart shape - rectangles, rounded rectangles, stadiums, circles, hexagons, decision diamonds - shows a + glyph on each of its four handles in edit mode. The + is the always-create affordance: drag from a + to start a new edge between two nodes. The handle area is now scoped tightly to the glyph itself, so clicking near a node corner (or on a diamond’s slanted edge) selects the node rather than accidentally starting a phantom edge.

Existing edges show small edge-update anchors at their endpoints when the edge is selected. Drag an anchor to re-route an existing edge to a different source or target - distinct from the + glyph, which always creates a new edge. Edge labels also no longer block clicks on the handle or the edge underneath them, so labelled edges are as easy to select and re-route as unlabelled ones.

Edge routing

Edges pick their routing style from the shape of the source node. A non-diamond parent (rectangle, rounded, stadium, circle, hexagon, etc.) uses junction routing: a plain bezier fan where every outgoing edge leaves from the same layout-default handle on the parent - arrows read left-to-right in LR layouts and top-to-bottom in TB layouts. A diamond parent uses decision routing: a custom staggered smoothstep edge that picks a distinct turn coordinate per sibling so option labels never overlap, even with five or more branches. Back-edges (targets behind the source on the primary axis) fall back to geometric handle selection in both modes so the loop-back is visually obvious.

Resizing and labels

Most div-based flowchart shapes can be resized in the visual editor. Decision diamonds auto-size from their label, keep text inside the visible dark diamond, and can be manually resized while preserving the square decision shape. Circles remain fixed-size. Subgraph / container nodes (swimlane lanes and grouped flowchart sections) are also manually resizable; touched groups keep their saved dimensions across regenerations, while untouched groups continue to receive fresh automatic sizing.

Version history

Every generation, every manual save, and every description change creates a new version - nothing is silently overwritten. Click the version badge (e.g. “v3 · Sep 2 · you”) to open version history.

• Load: restore any previous version.
• Delete: remove a version permanently.

Versions include the description, all diagram types, and creator information.

Sharing

Share your processes with colleagues via secure links.

1. Click the share button next to a process in the list.
2. Choose permission: Read (view only) or Write (can edit and generate).
3. Optionally set an expiration date.
4. Click Create link - the URL is copied to your clipboard.

Recipients need a Storm2Flow account in the same organization. You can revoke a share link at any time.

Spaces

A space is where a set of processes lives together. Every account has one private space (your own processes, only you can see them) and can join or create any number of shared spaces with colleagues from the same organisation.

The space switcher

The space switcher sits at the top-left of the header, next to the logo, showing the current space’s name and a coloured dot (grey for private, teal for shared). Click it to open a menu listing your private space and every shared space you belong to; selecting one scopes the process library on the left to that space and resets the workspace (any open process is closed so you are not looking at a stale view from another space).

Creating a shared space

From the space switcher menu, click Create shared space, type a name, and confirm. You become the space’s first owner. A new shared space starts empty - add processes by moving them in (next section).

Moving a process into a space

In your private space, top-level rows in the process library show a small ⇄ button next to the row actions (only when you have at least one shared space you can write to). Click it to pick a destination from the list of shared spaces where you are an owner or contributor; on pick, the whole process - including any sub-processes - moves into that space. The row leaves your private library and now appears in the chosen space’s library for every member.

To move a process back, open the shared space, find the row, click ⇄, and pick Private space. Only the owner of a process (the account that created it) can move it; other space members cannot move someone else’s content out.

Members & roles

When you have a shared space open and you are its owner, click the space switcher and pick Manage members to open the members panel. The panel lists every member with their email and role:

• Owner - read + edit + manage members (invite, change roles, remove).
• Contributor - read + edit processes in the space.
• Viewer - read only.

From the panel, an owner can invite any other active member of your organisation by email (the person must already have a Storm2Flow account - we do not send invitation emails to outsiders), change a member’s role, and remove a member. A space always keeps at least one owner - the last owner cannot be demoted or removed.

The invite field autocompletes colleagues from your organisation as you type, suggesting active teammates who are not already in the space (so you do not have to remember exact addresses, and you cannot accidentally re-invite someone). When you add a member, Storm2Flow emails them an invitation letting them know you added them to the space, with a link to open the app - so they find out without you having to ping them separately.

Leaving a space

Any member can leave a shared space via the Leave button in the members panel. On leave, the processes you contributed to that space (the ones you own and had moved in) return to your private space - nobody else inherits your work, and you keep your own content. Processes contributed by other members are unaffected and remain in the shared space.

Notes

• Spaces are scoped to your organisation; you cannot invite someone from a different org into a shared space, and a process never crosses an org boundary.
• The Add sub-process + As-is/To-be branching actions all work the same inside a shared space - the tree shape is independent of which space the root lives in.
• Per-process share links (Sharing, above) keep working alongside spaces. They grant a single recipient direct access to a single process, regardless of which space the process lives in.

Keyboard shortcuts

Shortcut	Action	Context
Ctrl+Scroll	Zoom in/out	Diagram canvas
← → ↑ ↓	Pan diagram	When not typing
Esc	Close modal	Any modal open
Enter	Confirm rename	Process name input

Mobile

Storm2Flow works on mobile devices with a tab-based navigation:

• Processes tab: browse and select processes.
• Editor tab: write descriptions, with a floating Generate button.
• Diagram tab: view diagrams, with a floating Export button.

All features work on mobile, including diagram editing, version history, and sharing.