Design

igly.ai

An MCP server for browsing templates and creating or editing Igly canvases with shapes, text, notes, lines, arrows, and freehand elements.

1 endpoint13 known toolsFirst detected October 18, 2026Last detected October 18, 2026

ENDPOINT 1

https://igly.ai/api/mcp

No auth detected

MCP server metadata

Name: igly-canvas
Version: 1.0.0

Capabilities

resources.listChangedtools.listChanged

Server instructions

Igly Canvas is a whiteboard you draw on with tools. - Start with create_canvas (blank, or template_id from list_templates). Keep the returned canvas_id. - For posters, social posts, menus, invitations, slides and other designed pieces, start from a template: list_templates shows the user a picker; create_canvas with template_id, or apply_template to add one to a board that already has content. Then edit the template's own text and shapes (get_canvas_summary lists their ids) instead of drawing it from scratch. - Coordinates are canvas units; x grows right, y grows down. A typical board fits in about 1600x1000. - Batch many elements in one add_elements call (up to 100). Give elements ids you can reuse in later calls. - Connect boxes with type "arrow" or "line" and from_id/to_id; connectors stay attached when boxes move. - When the user asks to change a board drawn here, edit that same canvas_id with these tools; never redraw it another way. - The user can also edit the board by hand. Read get_canvas_summary instead of guessing what is on it. Use export_png for an image of it. - The canvas opens in the full Igly editor through its open_url.

Known tools 13

list_templates

Browse published Igly Canvas templates (posters, social posts, menus, invitations...), featured first.

Inferred read-only

create_canvas

Create a new Igly canvas, blank or as an editable copy of a template.

Potential side effects

apply_template

Add a published template to an existing canvas.

Inferred read-only

get_canvas_summary

Compact list of what is on a canvas: one line per element (id type x,y wxh text), top layer first.

Inferred read-only

add_elements

Draw on a canvas: shapes (rectangle, ellipse, diamond, star...), sticky notes, text, lines and arrows (from/to points or from_id/to_id to connect elements), and freehand paths.

Inferred read-only

update_elements

Move (x/y or dx/dy), resize, rotate, recolor, retext, hide or lock elements by id.

Inferred read-only

remove_elements

Delete elements from a canvas by id.

Potential side effects

arrange

Align (left/center/right/top/middle/bottom), distribute evenly or with a gap, group/ungroup, or change stacking order.

Inferred read-only

export_png

Render the canvas (or some elements, or a region) to a PNG and return its link.

Inferred read-only

canvas_get_document

Full canvas document for the board view.

Inferred read-only

canvas_use_template

The board view's template picker: add a template to the open canvas, or start a new canvas from it when none is open.

Inferred read-only

canvas_get_revision

Current revision, so an open board view can reload after other changes.

Inferred read-only

canvas_save_document

Save the board view's edits against the revision it loaded.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.igly-canvas]
url = "https://igly.ai/api/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "igly-canvas": {
      "type": "http",
      "url": "https://igly.ai/api/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: igly-canvas
Remote MCP URL: https://igly.ai/api/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "igly-canvas": {
      "url": "https://igly.ai/api/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "igly-canvas": {
      "type": "http",
      "url": "https://igly.ai/api/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "igly-canvas",
  "transport": "streamable-http",
  "url": "https://igly.ai/api/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.