Skip to content
StudyPlug API

Overview

The Model Context Protocol (MCP) is an open standard that lets AI assistants call external tools and read external data. Instead of pasting API docs into a prompt, you give your AI agent a running MCP server and it discovers what tools are available, what parameters they accept, and how to call them.

The StudyPlug MCP server (@studyplug/mcp-server) exposes the full StudyPlug educational content engine as MCP tools. Any MCP-compatible client — Claude Desktop, Cursor, Windsurf, or a custom agent — can browse the K-5 curriculum, generate practice problems, and look up standards alignments without writing HTTP requests.

What the Server Provides

Section titled “What the Server Provides”

4 Tools

Section titled “4 Tools”
ToolPurpose
browse_curriculumDiscover grades, subjects, topics, and skills with progressive drill-down
generate_problemsGenerate practice problems for any skill with optional seed for reproducibility
find_by_standardLook up skills mapped to a CCSS or NGSS standard code
get_skill_infoGet detailed metadata about a skill including standards and grade availability

2 Resources

Section titled “2 Resources”
Resource URIContent
studyplug://curriculumFull curriculum tree as JSON (grades, subjects, topics, skills)
studyplug://standardsAll CCSS-Math, CCSS-ELA, and NGSS standards with mapped skills

How It Works

Section titled “How It Works”

The MCP server runs locally on your machine and communicates over stdio (standard input/output). When an AI agent connects:

  1. The agent calls listTools and discovers the 4 available tools.
  2. The agent reads the tool descriptions and parameter schemas to understand what each tool does.
  3. When the user asks something like “give me 10 multiplication problems for grade 3”, the agent calls generate_problems with the right parameters.
  4. The server runs the StudyPlug core generators directly (no HTTP round-trip to an external API) and returns structured text.

Because the server imports @studyplug/core directly, generation is fast and works fully offline.

When to Use MCP vs. the REST API

Section titled “When to Use MCP vs. the REST API”
  • Use MCP when an AI agent needs to interact with StudyPlug conversationally — tutoring sessions, lesson planning, adaptive practice.
  • Use the REST API when your application makes programmatic requests — building a worksheet app, quiz platform, or batch content pipeline.
  • Use both when your AI agent orchestrates a workflow that also feeds a UI (the agent uses MCP for discovery, then your app calls the REST API for rendering).

Next Steps

Section titled “Next Steps”