Play 74 — AI Tutoring Agent 🎓
Socratic AI tutor — adaptive difficulty, knowledge state tracking, misconception detection, personalized learning paths.
Build an intelligent tutoring agent that never gives direct answers. Uses Socratic questioning with 4-step hint progression, tracks student mastery via Bayesian Knowledge Tracing, detects and remediates misconceptions across sessions, and enforces strict content safety for minors.
Quick Start
cd solution-plays/74-ai-tutoring-agent
az deployment group create -g $RG -f infra/main.bicep -p infra/parameters.json
code .
# Use @builder to implement, @reviewer to audit, @tuner to optimizeArchitecture
Pre-Tuned Defaults
- Socratic: 4-step hint progression · 3 attempts before explanation · always check understanding
- Difficulty: 5 levels · advance after 3 correct streak · retreat on misconception
- Knowledge: Bayesian tracking · 0.80 mastery threshold · 14-day misconception decay
- Safety: All severity thresholds at 0 (zero tolerance for minors)
DevKit (AI-Assisted Development)
| Primitive | What It Does |
|---|---|
agent.md | Root orchestrator with builder→reviewer→tuner handoffs |
copilot-instructions.md | Tutoring domain (Socratic method, misconception handling, difficulty curves) |
| 3 agents | Builder (gpt-4o), Reviewer (gpt-4o-mini), Tuner (gpt-4o-mini) |
| 3 skills | Deploy (175+ lines), Evaluate (120+ lines), Tune (230+ lines) |
| 4 prompts | /deploy, /test, /review, /evaluate with agent routing |
Cost Estimate
| Service | Dev | Prod | Enterprise |
|---|---|---|---|
| Azure OpenAI | $30 | $350 | $1,500 |
| Cosmos DB | $3 | $95 | $360 |
| Azure AI Search | $0 | $250 | $500 |
| Static Web Apps | $0 | $9 | $9 |
| Container Apps | $10 | $120 | $350 |
| Azure AI Content Safety | $0 | $40 | $120 |
| Key Vault | $1 | $3 | $10 |
| Application Insights | $0 | $30 | $100 |
| Total | $44 | $897 | $2,949 |
vs. Play 65 (AI Training Curriculum)
| Aspect | Play 65 | Play 74 |
|---|---|---|
| Focus | Curriculum design + dependency graphs | Real-time Socratic tutoring |
| Interaction | Self-paced modules | Multi-turn adaptive conversation |
| AI Role | Generate learning paths | Guide student reasoning |
| Safety | Standard content safety | Strict (minors-grade, zero tolerance) |
📖 Full documentation · 🌐 frootai.dev/solution-plays/74-ai-tutoring-agent · 📦 FAI Protocol
FAI Manifest
| Field | Value |
|---|---|
| Play | 74-ai-tutoring-agent |
| Version | 1.0.0 |
| Knowledge | F1-GenAI-Foundations, O2-AI-Agents, T2-Responsible-AI |
| WAF Pillars | responsible-ai, reliability, cost-optimization, performance-efficiency |
Last updated on