Introduction

TutorFlow Agent Platform. AI-powered education APIs for automated grading, answer evaluation, and intelligent tutoring workflows.

What is Agent Platform?

TutorFlow Agent Platform is a set of REST APIs that let you integrate AI-powered education capabilities into your product. Grade essays, evaluate answers, and build intelligent education workflows, all with a single API call.

Key Features

  • AI Evaluation : Grade open-ended answers, rubric-based assessments, and exact-match questions with structured AI feedback.
  • Multi-Model Support : Backed by leading AI models. Switch between providers at any time without changing your integration code.
  • MCP Server : MCP-compatible AI agents can connect via the Model Context Protocol. No HTTP boilerplate needed.
  • Agent Self-Service : Autonomous agents can register, set up billing, and start evaluating without human intervention.
  • Webhooks : Real-time event delivery with automatic retries and HMAC signature verification.
  • LTI Integration : Connect directly to your LMS (Canvas, Moodle, Blackboard) via LTI 1.3 for seamless grade passback and assignment sync.
  • Multilingual : AI-generated feedback in 9+ languages.
  • Simple Pricing : One platform billing contract. Evaluation at $0.01, $0.03, or $0.05 per request by quality tier. No token math.

Quick Example

Send a single API request to evaluate a student's answer:

curl -X POST https://api.tutorflow.io/v1/platform/evaluations \
  -H "Authorization: Bearer tf_platform_..." \
  -H "Content-Type: application/json" \
  -d '{
    "evaluationType": "open_ended",
    "questionText": "Explain photosynthesis",
    "learnerAnswer": "Plants use sunlight to convert CO2 and water into glucose and oxygen.",
    "language": "en",
    "maxScore": 10
  }'

Two Ways to Integrate

REST API : Standard HTTP. Full control, works with any language or framework.

MCP Server : AI-native. Connect Claude Code or any MCP-compatible agent directly to TutorFlow tools without writing HTTP calls.

https://mcp.tutorflow.io/api/mcp

Next Steps