Phase 13

Tools and Protocols

Phase 13: Tools and Protocols. 23 hands-on lessons building AI from first principles in the browser. Free reading; graded exercises and certificate with lifetime access.

  1. The Tool Interface — Why Agents Need Structured I/O (graded)
  2. Function Calling Deep Dive — OpenAI, Anthropic, Gemini (graded)
  3. Parallel Tool Calls and Streaming with Tools (graded)
  4. Structured Output — JSON Schema, Pydantic, Zod, Constrained Decoding (graded)
  5. Tool Schema Design — Naming, Descriptions, Parameter Constraints (graded)
  6. MCP Fundamentals — Primitives, Lifecycle, JSON-RPC Base (graded)
  7. Building an MCP Server — Python + TypeScript SDKs (graded)
  8. Building an MCP Client — Discovery, Invocation, Session Management (graded)
  9. MCP Transports — stdio vs Streamable HTTP vs SSE Migration (graded)
  10. MCP Resources and Prompts — Context Exposure Beyond Tools (graded)
  11. MCP Sampling — Server-Requested LLM Completions and Agent Loops (graded)
  12. Roots and Elicitation — Scoping and Mid-Flight User Input (graded)
  13. Async Tasks (SEP-1686) — Call-Now, Fetch-Later for Long-Running Work (graded)
  14. MCP Apps — Interactive UI Resources via `ui://` (graded)
  15. MCP Security I — Tool Poisoning, Rug Pulls, Cross-Server Shadowing (graded)
  16. MCP Security II — OAuth 2.1, Resource Indicators, Incremental Scopes (graded)
  17. MCP Gateways and Registries — Enterprise Control Planes (graded)
  18. MCP Auth in Production — DCR, JWKS Rotation, Audience-Pinned Tokens on iii Primitives (graded)
  19. A2A — Agent-to-Agent Protocol (graded)
  20. OpenTelemetry GenAI — Tracing Tool Calls End-to-End (graded)
  21. LLM Routing Layer — LiteLLM, OpenRouter, Portkey (graded)
  22. Skills and Agent SDKs — Anthropic Skills, AGENTS.md, OpenAI Apps SDK (graded)
  23. Capstone — Build a Complete Tool Ecosystem
0 lifetime access. Curriculum based on AI Engineering from Scratch by Rohit Ghumare (MIT, used under attribution).