- Home
- Skills
- Technical Documentation
- StreamTeem Visual Page
Works with the AI Agents you already use
StreamTeem Visual Page
Complexity in architecture and data flows often hits a wall in text-only chat.
$1.99
StreamTeem Visual Page
Example session with this skill installed
Show me a sequence diagram of a user logging in, where the Client calls the Auth Service, which then queries the Database.
- Read your context and instructions
- Compiled the streamteem visual page
I have opened the visual page and generated the sequence diagram for the login flow. You can see the ordered calls from the Client to the Auth Service and the subsequent query to the Database.
Login Sequence Diagram
{
"language": "sequence",
"rev": 1,
"interactive": false,
"items": [
{ "id": "n1", "kind": "participant", "label": "Client" },
{ "id": "n2", "kind": "participant", "label": "Auth Service" },
{ "id": "n3", "kind": "participant", "label": "Database" },
{ "id": "e1", "kind": "call", "from": "n1", "to": "n2", "label": "POST /login" },
{ "id": "e2", "kind": "call", "from": "n2", "to": "n3", "label": "SELECT user" },
{ "id": "e3", "kind": "return", "from": "n3", "to": "n2", "label": "user_data" },
{ "id": "e4", "kind": "return", "from": "n2", "to": "n1", "label": "200 OK (JWT)" }
]
}
Next steps
- Review the sequence of calls on the visual page.
- If you need to add an MFA step, let me know and I will update the diagram.
- Press Save SVG on the page if you want to include this in your documentation.
Connects securely to your tools. The creator never sees your data.
About this skill
Complexity in architecture and data flows often hits a wall in text-only chat. This skill bridges the gap by launching a local, interactive visual whiteboard that syncs directly with your conversation.
What it does
- Multi-format diagramming Generates flowcharts, UML class models, ER diagrams, sequence diagrams, and timelines.
- Interactive editing Enables users to drag nodes, connect elements, rename labels, and circle items directly on the canvas.
- Bi-directional sync Reads user edits from the visual page and incorporates them into the next version of the diagram.
- Process animation Visualizes movement along routes using tokens to explain material flows or logic steps.
- Data visualization Converts numerical data from the chat into bar, line, or pie charts.
- SVG export Generates standalone SVG snapshots for inclusion in documentation or pull requests.
How it works
- Initialize server The agent launches a background Python process to serve the local whiteboard page.
- Publish scene The agent writes a JSON definition of the diagram to a local directory, which the page renders instantly.
- Review edits The agent monitors a JSON file for user interactions like moved nodes or new connections.
- Update & check The agent validates the updated scene using a CLI tool before pushing the revision to the browser.
Frameworks & tools
Uses a lightweight Python 3 server backend. Renders diagrams in any modern browser or agent preview pane. All data is stored locally in scene.json, edits.json, and snapshot.svg.
Why this beats prompting it yourself
Standard LLMs struggle to maintain consistent coordinate systems or complex UML relationships across multiple messages. This skill uses a structured JSON schema and a dedicated validation tool to ensure diagrams remain syntactically correct and editable, preventing the "broken syntax" common with raw Mermaid or SVG prompts.
Use cases
- Map out microservice architectures or network topologies.
- Design database schemas with keys and cardinality.
- Visualize complex API call sequences and timing.
- Track project milestones on an interactive timeline.
Known limitations
The Python server is local-only and defaults to port 8765. Requires a browser for viewing. Charts and sequence diagrams are read-only and do not support interactive dragging or relabeling.
Requires Python 3. Local-only server. Interactive editing is not supported for sequence, chart, or timeline formats. User must manually save SVGs via the browser UI.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 4 days ago
Requires
Permissions
Allowed Hosts
Starts a local Python server on 127.0.0.1 only, reads and writes whiteboard/*.json and snapshot.svg in the project, and opens the page in a browser. Network Access is for that local loopback server, not the public internet.