tesserae-mcp
Enables AI agents to build freeform dashboards for e-ink panels by listing widgets and devices, laying out a canvas, rendering a preview, and pushing to the panel.
README
tesserae-mcp
The MCP bridge for Tesserae. It lets an AI agent (Claude Desktop, Claude Code, or any MCP client) build freeform (canvas) dashboards for your e-ink panels: it lists your widgets and devices, lays out a canvas, renders a preview to check its own work, and pushes to a panel.
This is a thin stdio client. It talks to a running Tesserae over its /api/mcp
HTTP surface, so the rendering, widgets, and devices all come from your own
Tesserae instance.
create_canvas_page → set_canvas → render_preview → (look) → set_canvas → …
Prerequisites
A running Tesserae with the MCP API enabled: Settings → System → MCP → Enable MCP API. If this bridge runs on a different machine from Tesserae, also Regenerate token there and copy it.
Install
Run this on the machine where your agent runs (your laptop/desktop), which may be different from where Tesserae runs.
pip install git+https://github.com/dmellok/tesserae-mcp
That gives you the tesserae-mcp command.
Configure your agent
Point your MCP client at tesserae-mcp. Example (Claude Desktop / Claude Code
mcpServers config):
{
"mcpServers": {
"tesserae": {
"command": "tesserae-mcp",
"env": {
"TESSERAE_URL": "http://127.0.0.1:8765",
"TESSERAE_MCP_TOKEN": "<your-token>"
}
}
}
}
TESSERAE_URL— where your Tesserae is reachable (defaulthttp://127.0.0.1:8765). For a Docker/Home Assistant install, use its LAN address, e.g.http://192.168.1.50:8765(the port must be reachable; HA ingress-only setups won't work).TESSERAE_MCP_TOKEN— the token from Settings. Omit it when the agent and Tesserae share a machine (loopback is trusted).
Then just ask: "Build me an 800×480 dashboard with the time, today's weather for Melbourne, and my next calendar event, then show me a preview."
Tools
| Tool | What it does |
|---|---|
list_widgets |
Every placeable widget (with fragments) + theme/style/font options |
get_widget_options |
A widget's options + format hints (big choice lists omitted by default) |
get_widget_choices |
Page through one option's choice rows (HA entity pickers etc.) |
probe_widget_data |
A widget's data + data_source (live/sample/error) + bindable field paths |
list_devices |
Registered panels: dimensions + colour capability (palette, mono flag) |
list_pages |
Existing canvas dashboards |
create_canvas_page |
Create an empty canvas (size it to your panel) |
get_canvas |
Read a canvas document (returns a rev for concurrency-safe writes) |
set_canvas |
Replace a canvas document (422 with field errors if invalid) |
add_element |
Append one element (live-updates an open editor) |
update_element |
Change one element in place (no full re-send) |
delete_element |
Remove one element |
patch_canvas |
Change document-level fields (size, theme, bg) without touching elements |
arrange |
Compute aligned grid/row/column boxes so you lay out by intent, not pixels |
measure_text |
Measure rendered text width/height so a box fits its content |
render_report |
Read back what rendered (values, overflow, live-vs-sample, colours) as JSON |
render_preview |
Render the canvas to a PNG the agent can see |
push_to_device |
Push the canvas to explicit device(s) |
Writes (set_canvas, add_element, update_element, delete_element, patch_canvas)
accept an optional base_rev (from get_canvas); if the page changed since, the
write returns HTTP 409 so you re-read instead of clobbering a concurrent edit.
Run without installing
From a clone:
pip install mcp
TESSERAE_URL=http://127.0.0.1:8765 python -m tesserae_mcp
License
AGPL-3.0-or-later, matching Tesserae.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.