Agents

1.

Agentic Frameworks Overview WIP

High-level overview of multi-agent and code-centric AI frameworks

2.

Tools, Skills, & Connectors WIP

Function calling, skill libraries, plugins, and data connectors for AI Agents

3.

Agent Memory WIP

Types of agent memory (Procedural, Semantic, Episodic) and strategies for updating them

4.

Memory & Context Management WIP

Managing short/long-term memory and context windows in agents, including Mem0 and MemU frameworks

5.

Context Engineering & Filesystems WIP

How deep agents use filesystems as scratchpads for dynamic context engineering

6.

MCP (Model Context Protocol) WIP

Anthropic's Model Context Protocol for standardizing agent tool-use and data access

7.

Agentic RAG Frameworks WIP

Agentic RAG, self-reflection, iterative retrieval, and graph-based frameworks (LightRAG, RAG-Anything)

8.

Harness Engineering & Observability WIP

Evaluation harnesses, testing, and observability frameworks for agents (ART, Opik, AutoEvals)

9.

LangGraph WIP

Graph-based framework for building stateful multi-agent applications

10.

Agent Sandboxing WIP

Techniques and tools for safely executing agent-generated code

11.

Agent-to-Agent (A2A) WIP

Agent-to-Agent communication protocols

12.

The Ralph Wiggum Loop WIP

Understanding infinite failure loops and agent stagnation

13.

Claw (Open / Nemo) WIP

Open Claw and Nemo Claw frameworks/models for tool usage

14.

Lightweight Agents (nanobot) WIP

Minimalist, open-source AI agents (nanobot) for tools, chats, and workflows

15.

Hermes Agents WIP

Nous Hermes models and their agentic capabilities

16.

Agent Lightning WIP

Microsoft's extensible framework for RL-based training of any AI agent