mcp-server

Decantr AI MCP Server

Decantr brings frontend governance into AI coding workflows through 8 MCP tools. It serves developers and coding assistants working with project contracts, route-specific context, local verification evidence, and scoped repairs for AI-edited interfaces.

Server facts

mcp-server

Tech

Runtime / language
Not specified
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
Not recordedPart of the decantr-ai/decantr monorepo
Last commit
Data as of

Overview

Frontend changes guided by AI can be checked against Decantr’s project contracts and local evidence through its MCP interface. The project tool exposes app discovery, adoption status, governing rules, and workspace health without sending source code elsewhere. Contract operations cover reading, validation, drift inspection, and compatibility skeleton creation. Context preparation focuses assistants on the relevant UI surface and route, while graph queries connect that work to local source evidence. Together, these tools support assistants working within an existing project’s rules rather than relying only on the latest prompt.

Verification examines local UI changes and can provide critique, findings, health information, and evidence bundles. The repair tool uses typed findings to prepare bounded repair plans, prompts, and guidance for the health loop. Contract changes have a separate, explicit write tool limited to the workspace, including accepting drift and updating the Essence v4 contract. A compatibility tool named decantr_registry provides access to Decantr’s official content collection, benchmark information, and execution packs; it does not expose a public registry. The server uses stdio transport.

mcp-server setup notes

  • mcp-server records stdio only; confirm a network transport upstream before planning remote access.
  • mcp-server: Part of the decantr-ai/decantr monorepo
  • mcp-server has no recorded license; confirm reuse terms in its repository.

Related MCP servers