About Foundations Elements
Foundations Elements is a remote catalog server for agents and design-oriented workflows that need to find UI elements and retrieve their specifications. This package provides the client plugin and registry metadata; access to catalog tools requires a Foundations account.
Key features
- 01Search the Elements catalog
- 02List element kinds
- 03Fetch component specifications
- 04Connect through a remote HTTPS endpoint
Use cases
- 01Find UI elements for a design task
- 02Retrieve specifications for a component
- 03Browse available element kinds
MCP support
The remote MCP server exposes catalog search, kind-listing, and specification-fetch tools. MCP clients connect over HTTPS and authenticate with Foundations OAuth.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://foundations.tensegrity.works/mcpclaude mcp add --transport http works-tensegrity-foundations-elements https://foundations.tensegrity.works/mcp{"mcpServers":{"works-tensegrity-foundations-elements":{"type":"http","url":"https://foundations.tensegrity.works/mcp"}}}Open in Cursor{"servers":{"works-tensegrity-foundations-elements":{"type":"http","url":"https://foundations.tensegrity.works/mcp"}}}Open in VS Code{"mcpServers":{"works-tensegrity-foundations-elements":{"type":"http","url":"https://foundations.tensegrity.works/mcp"}}}{"mcpServers":{"works-tensegrity-foundations-elements":{"serverUrl":"https://foundations.tensegrity.works/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEFoundations Elements MCP
Public install package for the Foundations Elements remote MCP server.
- MCP URL: https://foundations.tensegrity.works/mcp
- Product / waitlist: https://tensegrity.company/elements
- Docs: https://foundations.tensegrity.works/docs
- Privacy: https://tensegrity.company/privacy
- Terms: https://tensegrity.company/terms
- Support: [email protected]
This repository is the free plugin / registry metadata. The Worker implementation stays private. Access to catalog tools requires a Foundations account (OAuth via https://auth.tensegrity.works).
Publisher (marketplaces): GitHub org Tensegrity-Systems. Canonical repo: https://github.com/Tensegrity-Systems/foundations-elements-mcp (GitLab remains a staging mirror).
What you get
Cursor / Agent plugin that points at the remote MCP endpoint, plus server.json for the official MCP Registry.
Tools include search, kind lists, and spec fetch (elements_search, elements_list, elements_get, and related). See server instructions after connect.
Install
Cursor
- Install this plugin from the Marketplace when published, or load the repo as a local plugin.
- Complete OAuth when Cursor prompts for Foundations sign-in.
- Request access at https://tensegrity.company/elements if you do not have an account.
mcp.json already points at:
https://foundations.tensegrity.works/mcp
Claude
Add a custom connector with the same HTTPS URL, then sign in with OAuth. Directory listing will use this package’s branding and docs links.
VS Code / Copilot
Configure an HTTP MCP server with the URL above (after Official / GitHub MCP Registry publication).
Official MCP Registry
server.json declares a remote Streamable HTTP server. Preferred registry name (after domain ownership proof on tensegrity.works):
works.tensegrity/foundations-elements
Privacy
Foundations account sign-in and MCP usage are covered by the company privacy policy:
https://tensegrity.company/privacy
This install package does not collect data on its own; it only points clients at the remote MCP URL. Support: [email protected].
License
MIT for this install package. Foundations Elements content and API remain proprietary; see https://tensegrity.company/terms and https://foundations.tensegrity.works/terms.
Channel note
Paid HTTP (x402) lookups use GET /mcp/elements/search. That is a separate agent-catalog path, not this IDE MCP plugin.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.