Introduction

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, generate teaching content, and build education workflows with API calls.

Key Features

• Evaluation: Grade open-ended answers, rubric-based assessments, and exact-match questions with structured feedback. Open-ended and rubric evaluation use AI; exact-match evaluation is deterministic.
• Course Generation: Turn a topic prompt into a structured curriculum with chapters, lessons, lectures, and quizzes in one call.
• Module Generation: Generate one reusable interactive lesson with the same content shape as TutorFlow classroom modules.
• Slide Generation: Generate a deck with editor content, optional AI images, and built-in PowerPoint export.
• Video Generation: Produce narrated short-form videos with script, TTS, stock clips, subtitles, transitions, BGM, and optional Remotion-based render to mp4.
• Managed AI Runtime: TutorFlow chooses and operates the AI providers behind each API. Public requests do not need model or provider parameters.
• MCP Server: MCP-compatible AI agents can connect via the Model Context Protocol for onboarding, billing, pricing, evaluations, and courses. Use REST for modules, tests, slides, and videos.
• Agent Self-Service: Autonomous agents can register and start generating with trial credits. Paid top-ups use Stripe checkout or x402 when enabled.
• 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 content and AI-backed evaluation feedback in 9+ languages.
• Simple Per-Unit Pricing: One platform billing contract. Pay only for what you generate: $0.02/evaluation, $0.05/lesson, $0.05/module, $0.04/scene, $0.03/slide, $0.015/test item. No tiers, no token math.

Pick Your API

Generation APIs share the same operational patterns, but their response shapes are intentionally different:

• Courses, tests, slides, and videos return shareable URLs. Courses, tests, slides, and videos also support editor token flows.
• Modules return a publicUrl and direct module fields, but no edit token.
• Evaluations return grading results and learner mastery data, not content URLs.
• Synchronous mode returns the result in 15-90 seconds where mode is supported.
• Async mode (mode: "async") returns immediately and you poll or listen for a webhook.
• Create endpoints support idempotency with Idempotency-Key where documented.

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.",
    "language": "en",
    "submitter": {
      "learnerId": "9c4d2e1f",
      "learnerName": "Alex"
    }
  }'

Or generate a slide deck:

curl -X POST https://api.tutorflow.io/v1/platform/slides \
  -H "Authorization: Bearer tf_platform_..." \
  -H "Content-Type: application/json" \
  -d '{ "prompt": "Mediterranean diet overview", "slideCount": 6 }'

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

• Follow the Quick Start guide to make your first API call.
• Connect an AI agent via the MCP Integration guide.
• Learn about Agent Self-Service Onboarding for fully autonomous agents.
• Learn about Authentication and API key management.
• Review Pricing and Billing before going live.
• Explore the API Reference for detailed endpoint documentation.