The RoboNuggets Community

Agentic AI
Foundations

Tools change weekly; the mental model behind every AI agent doesn't. Understand the model, the harness, and context - then build the skills and command centre to run a whole team of them.

5 volumes·48 modules·Community-exclusive
1
VOLUME 1 · ANATOMY

The Anatomy of an AI Agent

Tools change weekly. The mental model behind every AI agent doesn't. Get this right once and every new launch clicks.

Module 1

The Mental Model

  • Pace of change
  • The durable pattern
01
Module 2

Chatbots vs Agents

  • How chatbots work
  • How agents differ
02
Module 3

The Model

  • Inside an LLM
  • Opus / GPT / Gemini
03
Module 4

The Harness

  • Inside a harness
  • Claude Code / Codex / Antigravity
04
Module 5

The Context

  • Plain English instructions
  • CLAUDE.md and AGENT.md
05
Module 6

The Agentic Loop

  • Look, think, act, check, repeat
  • Stop conditions
06
2
VOLUME 2 · DEEP DIVE

The Model, the Harness, and Context

Going deeper on the three parts. What each one actually does - and how they fit together.

Module 7

The 6-Month Rule

  • Plan for next-gen models
  • Spikiness explained
07
Module 8

Model Tiers

  • High / mid / low
  • Anthropic / OpenAI / Google
08
Module 9

Tokens

  • Tokens explained
  • Bible benchmark
09
Module 10

Context Window

  • Working memory
  • Overflow problem
10
Module 11

Harness Surfaces

  • Terminal, desktop, IDE, messaging
  • IDE = the sweet spot
11
Module 12

CLAUDE.md and AGENT.md

  • The agent's job description
  • Pre-injected every session
12
Module 13

Skills 101

  • Inside a skill
  • One file, one workflow
13
Module 14

Memory 101

  • How memory works
  • MEMORY.md index pattern
14
3
VOLUME 3 · LANGUAGE

The Language of Agents

Demystify the jargon. Every term you'll see in docs, tweets, and tutorials - in plain English.

Module 15

IDEs

  • IDEs explained
  • VS Code / Cursor / Antigravity
15
Module 16

Permission Modes

  • Auto / Ask / Plan / Bypass
  • When to use each
16
Module 17

Tool Permissions

  • Per-project tool rules
  • Allow / ask / deny
17
Module 18

Effort

  • How effort works
  • Token implications
18
Module 19

Model Picker

  • Opus / Sonnet / Haiku
  • When to switch
19
Module 20

Model Usage

  • Token spend tracking
  • Quotas and rate limits
20
Module 21

Context Hygiene

  • Token bloat problem
  • Pruning techniques
21
Module 22

Compact & Clear

  • What /compact does
  • What /clear does
22
Module 23

Skill Patterns

  • Trigger conventions
  • Frontmatter + body
23
Module 24

Memory Layers

  • Index file and individual files
  • External systems
24
Module 25

MCP Servers

  • MCP explained
  • Gmail / Drive / Slack / Linear
25
Module 26

CLI Tools

  • Bash / git / gh
  • Built-in tool calls
26
Module 27

Plugins

  • How plugins work
  • Custom slash commands
27
Module 28

Routines

  • How routines run
  • Cron syntax basics
28
Module 29

Hooks

  • How hooks fire
  • Pre / post triggers
29
4
VOLUME 4 · SKILLS & ORCHESTRATION

Core Agentic and Orchestration Skills

The skills that turn one agent into a team. Build your own, calibrate as you go, run multiple sessions on one project.

Module 30

Build Your Skills

  • Skill anatomy
  • master-* prefix convention
30
Module 31

Calibrate

  • End-of-session ritual
  • Light / full / auto modes
31
Module 32

Coordinate

  • Shared project folder
  • context.md + session-log
32
Module 33

Onboard

  • Pick up any project
  • Brief before effort
33
Module 34

Align

  • Clarifying questions first
  • Scope before effort
34
Module 35

Devil

  • Contrarian view
  • Break the yes-man pattern
35
Module 36

Burst

  • N variations on demand
  • Pick the winner
36
Module 37

Tweak

  • Live UI sliders
  • Bake the chosen design
37
Module 38

Session Continuity

  • Convo logs and handoffs
  • Fork conversation
38
Module 39

Multi-Agent Orchestration

  • When to split
  • Why specialists win
39
5
VOLUME 5 · COMMAND CENTRE

Your AI Command Centre

Every agent. Full visibility. The dashboard layer nobody else teaches.

Module 40

The Why

  • Visibility at scale
  • The 3-agent threshold
40
Module 41

The Dashboard

  • Single-screen overview
  • Live agent state
41
Module 42

Agent Statuses

  • Idle / working / blocked
  • Live polling
42
Module 43

Sprint & Backlog

  • The sprint board
  • The backlog queue
43
Module 44

Generations

  • Image / video / audio
  • Cost per generation
44
Module 45

Docs

  • Living markdown docs
  • Agent read + write
45
Module 46

Flows

  • Multi-step pipelines
  • Schedule or trigger
46
Module 47

Skill Tree

  • Skill map per agent
  • Spotting duplication
47
Module 48

Building Your Own

  • Open-source starter
  • Or fork ours
48