From fcc3830226561b8c83cb25f3c9eb71680c480e34 Mon Sep 17 00:00:00 2001 From: Shayan Rais Date: Fri, 24 Apr 2026 21:18:45 +0500 Subject: [PATCH] skills --- .../index.html | 200 +++++++++++++++--- 1 file changed, 175 insertions(+), 25 deletions(-) diff --git a/presentation/2026-04-25-gdg-kolachi-cli-claude-code-gemini/index.html b/presentation/2026-04-25-gdg-kolachi-cli-claude-code-gemini/index.html index 47ecd95..698a89b 100644 --- a/presentation/2026-04-25-gdg-kolachi-cli-claude-code-gemini/index.html +++ b/presentation/2026-04-25-gdg-kolachi-cli-claude-code-gemini/index.html @@ -2019,6 +2019,7 @@ todoapp/

The model's working memory — what it can see in this moment.

+ ✅ Fresh every chat ⚠️ Dumb-zone problem ⚠️ /compact
@@ -2062,19 +2063,168 @@ todoapp/ + +
+
+

🧠 Claude's Brain

+
+

Think of it like this

+

Imagine Claude has a brain that holds everything it's aware of right now — your question, every file it's opened, every tool result, every word it's said back to you. If a thought isn't in the brain, Claude can't use it. Simple as that.

+
+ Context window diagram showing how the 1M token limit is divided across system prompt, tools, files, and conversation +
+
+ 💬 +
+ Your messages + Every question, clarification, and follow-up +
+
+
+ 📄 +
+ Every file Claude opens + Read a 500-page PDF? All of it is in the brain now. +
+
+
+ 🔧 +
+ Every tool result + Web searches, command output, screenshots — all held in memory +
+
+
+
+

Two Things to Know

+

1. The brain is finite. It can hold about 1 million tokens — roughly 750,000 words. Big, but not infinite. 2. The brain empties at the end of every chat. When you start a new conversation, Claude remembers nothing from the last one unless you tell it again.

+
+
+
+
+ 🧑‍💼 + Agents + — the specialists + A dedicated Claude worker — own context, own tools, own focus. Each runs in isolation. + ✅ fresh working memory per run +
+
+ 📝 + CLAUDE.md + — your memory + Knowledge you provide to the model. Read every session. Keep it under 200 lines. + ⚠️ 200-line problem +
+
+ 🎯 + Skills + — the know-how + What the specialist (or Claude) can actually do. Loaded on demand — only when needed. + ✅ progressive disclosure +
+
+ 💭 + Context + — the working memory + What Claude holds in his head now — fresh every new chat. Finite budget; compact at ~50%. + ⚠️ dumb-zone problem +
+
+ 📘 + Workflows + — the instruction manual + Repeatable step-by-step recipes — like an AC install guide. Deterministic, reproducible. + ✅ reproducible recipes +
+
+
+ + +
+
+

What Loads at Session Start

+

The moment you open Claude Code, certain things land in Claude's brain before you've typed a word. The rest waits in the wings — only loaded when you actually need it. This is called progressive disclosure.

+ Diagram showing what loads into Claude's context window at session start +
+
+ 📋 +
+ CLAUDE.md — full content + Standing instructions, every line. Always pinned in the brain at startup. +
+
+
+ 🎓 +
+ Skills — descriptions only + Claude knows which skills exist and what each does — the full content loads when the skill is actually invoked. +
+
+
+ 👤 +
+ Agents — descriptions only + The roster of specialists. The full system prompt loads when the agent is dispatched. +
+
+
+
+

The One-Liner

+

Only descriptions of skills and agents are loaded at startup — the rest is fetched on-demand. That's progressive disclosure. It keeps the brain light.

+
+
+
+
+ 🧑‍💼 + Agents + — the specialists + A dedicated Claude worker — own context, own tools, own focus. Each runs in isolation. + ✅ fresh working memory per run +
+
+ 📝 + CLAUDE.md + — your memory + Knowledge you provide to the model. Read every session. Keep it under 200 lines. + ⚠️ 200-line problem +
+
+ 🎯 + Skills + — the know-how + What the specialist (or Claude) can actually do. Loaded on demand — only when needed. + ✅ progressive disclosure +
+
+ 💭 + Context + — the working memory + What Claude holds in his head now — fresh every new chat. Finite budget; compact at ~50%. + ⚠️ dumb-zone problem +
+
+ 📘 + Workflows + — the instruction manual + Repeatable step-by-step recipes — like an AC install guide. Deterministic, reproducible. + ✅ reproducible recipes +
+
+
+ -
+
Topic 2

🎓 Skills — What the Weather Reporter Knows

Skills are the specific things the reporter has been trained to do. Our reporter has two: fetch the data, and render it as a card.

-
+

The Training Manual

Think of it like this

@@ -2101,7 +2251,7 @@ todoapp/
-
+

When to Turn Something Into a Skill

Tip from Boris Cherny (creator of Claude Code) — Feb 1, 2026

@@ -2138,7 +2288,7 @@ todoapp/
-
+

Why Separate Agents and Skills?

@@ -2164,7 +2314,7 @@ todoapp/
-
+

How to Create Your Own Skill

Skills are plain markdown files. If you can write a recipe, you can write a skill.

@@ -2217,7 +2367,7 @@ Fetch the current temperature for Dubai, UAE.
-
+

Skill Config Fields

The small config block at the top of a SKILL.md (the "frontmatter") controls how the skill behaves:

@@ -2257,14 +2407,14 @@ Fetch the current temperature for Dubai, UAE. -
+
Topic 3

🧠 Context — The Reporter's Brain

Now that you've met the reporter and know their skills, let's understand what they can actually hold in mind at once. Every agent — including the weather reporter — gets its own fresh brain.

-
+

Claude's Brain

Think of it like this

@@ -2302,7 +2452,7 @@ Fetch the current temperature for Dubai, UAE.
-
+

What Loads at Session Start

The moment you open Claude Code, certain things land in Claude's brain before you've typed a word. The rest waits in the wings — only loaded when you actually need it. This is called progressive disclosure.

Diagram showing what loads into Claude's context window at session start @@ -2336,7 +2486,7 @@ Fetch the current temperature for Dubai, UAE.
-
+

Keep the Brain Clear

The more stuff crammed into Claude's brain, the harder it is to focus on what matters. This is called context rot — performance drops as the brain gets crowded.

@@ -2380,7 +2530,7 @@ Fetch the current temperature for Dubai, UAE.
-
+

How to Manage Your Context

You can't create the context — it's just there, the moment you open a chat. But you can see how full it is, trim it down, or wipe it clean. Three commands give you full control.

@@ -2425,14 +2575,14 @@ Fetch the current temperature for Dubai, UAE. -
+
Topic 4

📋 CLAUDE.md — The Reporter's Pocket Rulebook

The weather reporter consults this at the start of every shift — even though their brain resets overnight. It's the standing instructions pinned in that brain before you've said a word.

-
+

The Employee Handbook

Think of it like this

@@ -2460,7 +2610,7 @@ This is the Q2 marketing campaign brief — targeting small business owners.
-
+

How to Create Your CLAUDE.md

You don't need to write CLAUDE.md by hand. Claude can look at your project and draft one for you.

@@ -2505,7 +2655,7 @@ open, edit, save # Tweak it like any doc<
-
+

Grow CLAUDE.md With Every Mistake

Tip from Boris Cherny (creator of Claude Code) — Feb 1, 2026

@@ -2542,7 +2692,7 @@ open, edit, save # Tweak it like any doc<
-
+

What Goes in CLAUDE.md

# CLAUDE.md @@ -2570,7 +2720,7 @@ This is a TodoApp with a FastAPI backend and React frontend.
-
+

How CLAUDE.md Loads

Claude Code uses two mechanisms to find CLAUDE.md files:

@@ -2603,14 +2753,14 @@ This is a TodoApp with a FastAPI backend and React frontend. -
+
Topic 5

⚡ Commands — The Trigger

One word kicks off the whole chain. /weather-orchestrator → agent → skill → SVG card. Commands are the entry point into any workflow.

-
+

Commands — The Entry Point

Think of it like this

@@ -2654,7 +2804,7 @@ This is a TodoApp with a FastAPI backend and React frontend.
-
+

How to Create Your Own Command

Commands are markdown files too. If you can write a recipe, you can write a command.

@@ -2709,14 +2859,14 @@ This is a TodoApp with a FastAPI backend and React frontend. -
+
Putting It All Together

🎼 Workflow — All Five Pieces Together

Watch the weather reporter example run from one keystroke to SVG card output. Five concepts, one orchestrated flow.

-
+

Command → Agent → Skill

This is the core architecture pattern of Claude Code workflows — demonstrated in this very repo by the weather example:

The Orchestration Flow @@ -2743,7 +2893,7 @@ This is a TodoApp with a FastAPI backend and React frontend.
-
+

Two Ways Skills Are Used

The weather workflow demonstrates both skill patterns in a single flow:

@@ -2766,7 +2916,7 @@ This is a TodoApp with a FastAPI backend and React frontend.
-
+

How to Wire Your Own Workflow

A workflow isn't a separate file type. It emerges when one command calls agents and skills in sequence.

@@ -2817,7 +2967,7 @@ This is a TodoApp with a FastAPI backend and React frontend. -
+

Journey So Far

Five concepts, one running example

From meeting the weather reporter to wiring the full Command → Agent → Skill chain. The same five pieces compose every workflow you'll ever build.