MCP server
An MCP server for AI coding agents
Schemyx MCP gives AI coding agents a standard bridge to your product context. Instead of pasting rules into every prompt, agents can ask the MCP server for the exact recipe they need.
The problem
Agents need a reliable bridge to project knowledge.
Without a tool interface, project rules live in prompts, docs, and local memory. Agents have to guess where the current source of truth lives.
Schemyx approach
Schemyx MCP turns project knowledge into callable context.
The MCP server can read local Schemyx bundles or hosted project configs and return focused recipes for styles, components, routes, APIs, services, models, clusters, and rules.
Why teams use it
Less drift, faster onboarding, and the same context across agents.
Outcome
Works where developers already build
MCP brings Schemyx context into AI coding tools instead of forcing developers to leave the workflow.
Outcome
Local-first context
Teams can keep bundles in the repo for portability and use hosted configs when they need shared account-backed workflows.
Outcome
Precise responses for lower token overhead
An MCP lookup can return a tiny recipe or context pack instead of a pile of unrelated code.
Lookup examples
The useful detail lives behind stable keys.
Schemyx is built for exact reads. Agents can ask for a concept, resolve a key, then fetch a tiny context response before they write.
schemyx_search_codebase
Search the context layer
Resolve natural requests like button styles or beta access API into stable keys.
schemyx_get_codebase_recipe
Fetch one recipe
Return the exact component, style, route, API, service, model, or rule context.
schemyx_get_context_pack
Fetch related context
Return the resolved recipe with dependencies, used-by context, and review items.
schemyx_read_codebase_file
Fallback source read
Read the source file only when the recipe is not enough.
Workflow
From codebase scan to agent-safe generation.
Schemyx keeps the heavy context work out of the prompt and turns repeated decisions into reusable lookup responses.
01
Install the MCP server
Connect your AI coding tool to Schemyx local context.
02
Generate or load config
Use local bundles, hosted project configs, or auto mode.
03
Call lookup tools
Agents search, fetch recipes, request packs, and inspect review items.
04
Write with context
Generated changes start from known product rules instead of guesses.
Who it is for
Built for teams and builders who want AI work to stay coherent.
Audience
AI-heavy engineering teams
Give every agent access to the same current project context.
Audience
Local-first developers
Use committed bundles without sending source code to a hosted scanner.
Audience
Teams standardizing workflows
Make MCP the delivery layer for approved rules and codebase intelligence.
FAQ
Questions teams ask before trusting agent context.
What does the Schemyx MCP server serve?
It can serve theme bundles, component rules, and local codebase recipes for files, components, routes, APIs, services, models, styles, clusters, and review items.
Does MCP write code?
No. Schemyx MCP serves context. The coding agent decides how to apply that context inside the repo.
Can it work without hosted configs?
Yes. Local bundles can live inside the repo, and auto mode can prefer local files before hosted project configs.
Why not paste the same rules into prompts?
Prompts drift and get expensive. MCP gives agents a repeatable way to fetch the current rule when they need it.
Next step
Give every agent the same source of truth.
Request beta access or book a call to map Schemyx against your current stack, team standards, and agent workflow.