workflow added
This commit is contained in:
Shayan Rais
@@ -181,3 +181,4 @@ After completing changes, report to the user:
|
||||
- **2026-04-24 Skills emoji updated + 2 new slides inserted at 38-39 (63 → 65 slides)**: Edit 1 — slide 37's h1 emoji changed from `🎓` (graduation cap) to `🎯` (target) to match the Skills pillar-footer card emoji established on slide 14 and used across all 12 pillar footers. Edit 2 — new slide 38 ("Create your first skill", `data-level="skills"`) inserted with a `.two-col` layout: two `.col-card` divs (both green `#4caf50` border-left) for Method 1 (write by prompting) and Method 2 (use Anthropic's `skill-creator`), followed by an inline amber warning box (`background: #fff3e0; border-left: 4px solid #ff9800`) flagging the wrong-structure gotcha of Method 1. Edit 3 — new slide 39 ("Skill config fields with frontmatter", `data-level="skills"`) mirrors slide 29's agent field-table pattern exactly: 8 field rows using `.field-row` / `.field-name` / `.field-desc` / `.field-enforced` / `.pill-harness` / `.pill-prompt` CSS classes, italic intro sentence, and a centered `.fafafa` footnote callout for `description`. Old slides 38-63 renumbered to 40-65 via Python sentinel replacement (descending-order sentinel phase + resolve phase). **Redundancy flag**: old Skills section opener (was slide 39, now slide 41 — `data-level="skills"`, `section-slide` class, "Topic 2 — What the Weather Reporter Knows") is now downstream of the new teaching slides 38-39 which already fire `→ skills` earlier. Same pattern as the CLAUDE.md dual-fire situation at slides 33 and 52. Recommended: keep slide 41 as a breathing-space divider into the weather-reporter narrative arc; the journey bar will not re-fire (same level). Final section-divider positions: **Agents=23, Skills=38 (first skills data-level), Context=47, CLAUDE.md=52, Commands=58, Workflow=61**. Total slides: **65**.
|
||||
- **2026-04-24 context-section concept-intro enrichment (65 → 68 slides)**: (1) Green pill `✅ Fresh every chat` (bg `#2e7d32`) added to slide 41 as the first pill before the 2 existing red pills — visual order green/property first, red/pitfalls after. (2) New slide 42 ("Claude's Brain", `data-level="context"`) and slide 43 ("What Loads at Session Start", `data-level="context"`) inserted between slide 41 and the old Skills section opener (old 42 → 44). Both new slides pull content forward from the old deep-dive slides 49/50 (which become slides 51/52 after the +2 renumber). Both carry full pillar-footers with Context active. Renumber technique: Python sentinel with per-occurrence targeting — `data-slide="42" data-level="skills"` and `data-slide="43">` were each targeted individually before the bulk 44-66→46-68 sweep, because slides 42/43 now have two occurrences each (new context + old skills). The CSS rule `.slide[data-slide="1"]` creates a false-positive when counting `data-slide` occurrences via regex — use `Counter` + ignore the CSS line, or grep for `<div class="slide"` instead. **Redundancy flag on slides 51 and 52**: old "Claude's Brain" (slide 51) and "What Loads at Session Start" (slide 52) are now duplicated by new slides 42 and 43. The old versions sit inside the deep-dive context section (slides 50-54) and will still fire at level=context (they carry no explicit `data-level`, but the section divider at slide 50 fires context and all subsequent slides inherit it until the next level fires). Recommend demoting or removing old 51/52 after presenter confirms new slides are satisfactory. Final section-divider positions: **agents=23, skills=38, context=42 (new), claude-md=55, commands=61, workflow=64**. Total slides: **68**.
|
||||
- **2026-04-24 "Lost in the Middle" research slide inserted at position 44 (68 → 69 slides)**: new slide inserted between slide 43 ("What Loads at Session Start") and old slide 44 (Skills section divider, now 45). Layout: flex two-column — image left (`max-width: 420px`, placeholder `../assets/concepts/context/lost-in-middle.png`), explanation right (`<ol>` with 3 items). Yellow callout div (`background: #fff8e1; border-left: 4px solid #f9a825`) at bottom of right column ties the academic finding to the "dumb-zone problem" vocabulary already established. Image file did NOT exist at insertion time — placeholder path used; user must drop `lost-in-middle.png` into `presentation/assets/concepts/context/`. Renumber method: Python sentinel replacing `data-slide="44"` through `"68"` → `"45"` through `"69"`, with a first-occurrence restore for the new slide's own `"44"` sentinel. New section-divider positions: **agents=23 (unchanged), skills=45, context=51, claude-md=56, commands=62, workflow=65**. Total slides: **69**.
|
||||
- **2026-04-24 mass deletion (slides 45-69) + single replacement + image relocation (69 → 45 slides)**: (1) Images `context-window.jpeg` and `context.jpg` moved from `presentation/assets/concepts/` to `presentation/assets/concepts/context/` alongside the already-resident `lost-in-the-middle.png`. All 5 `<img src>` references updated via sed (2 old-path `context-window.jpeg` refs, 2 old-path `context.jpg` refs, 1 `lost-in-middle.png` → `lost-in-the-middle.png` filename fix). (2) Slides 45-69 deleted (all 25 slides + their banner comments, lines 2289-3057 in 1-indexed) by slicing the lines array in Python and re-joining — the Python line-slice approach is the cleanest for deleting a contiguous range; no sentinel needed because no renumbering occurs within the retained block. (3) New slide 45 inserted in the same atomic write: `section-slide` with `data-level="workflow"`, h1 `📘 Workflow`, yellow pill `.claude/commands/`, green pills for Reproducible recipes / Same steps / Team-shareable, and a full pillar-footer with Workflows as the ONLY active card (all four others inactive). **Final state**: 45 slides, contiguous 1-45. data-level assignments: agents=23 (first), claude-md=33 (first), skills=38 (first), context=42 (first), workflow=45. Note: `commands` level is gone from data-level assignments (no remaining command-section slides). `LEVEL_LABELS` in JS still contains `'commands'` as a key — this is harmless (no slides will fire it) but should be removed in a future cleanup pass if desired. **Python line-slice pattern**: `before = lines[:del_start_0idx]`, `after = lines[del_end_0idx + 1:]`, new_content = join(before) + new_html + join(after). Atomic, no collision risk, trivially auditable.
|
||||
|
||||
@@ -2071,7 +2071,7 @@ todoapp/
|
||||
<h4>Think of it like this</h4>
|
||||
<p>Imagine Claude has a <strong>brain</strong> 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.</p>
|
||||
</div>
|
||||
<img src="../assets/concepts/context-window.jpeg" alt="Context window diagram showing how the 1M token limit is divided across system prompt, tools, files, and conversation" style="width: 100%; max-width: 800px; margin: 24px auto; display: block; border-radius: 8px;" />
|
||||
<img src="../assets/concepts/context/context-window.jpeg" alt="Context window diagram showing how the 1M token limit is divided across system prompt, tools, files, and conversation" style="width: 100%; max-width: 800px; margin: 24px auto; display: block; border-radius: 8px;" />
|
||||
<div class="use-cases">
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">💬</span>
|
||||
@@ -2144,7 +2144,7 @@ todoapp/
|
||||
<div style="min-height: calc(100vh - 120px);">
|
||||
<h1>What Loads at Session Start</h1>
|
||||
<p>The moment you open Claude Code, certain things land in Claude's brain <strong>before you've typed a word</strong>. The rest waits in the wings — only loaded when you actually need it. This is called <strong>progressive disclosure</strong>.</p>
|
||||
<img src="../assets/concepts/context.jpg" alt="Diagram showing what loads into Claude's context window at session start" style="width: 100%; max-width: 800px; margin: 24px auto; display: block; border-radius: 8px;" />
|
||||
<img src="../assets/concepts/context/context.jpg" alt="Diagram showing what loads into Claude's context window at session start" style="width: 100%; max-width: 800px; margin: 24px auto; display: block; border-radius: 8px;" />
|
||||
<div class="use-cases" style="margin-top: 24px;">
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">📋</span>
|
||||
@@ -2221,7 +2221,7 @@ todoapp/
|
||||
<div style="display: flex; gap: 40px; align-items: flex-start; flex-wrap: wrap;">
|
||||
<!-- Left column: image -->
|
||||
<div style="flex: 0 0 auto; display: flex; align-items: center; justify-content: center;">
|
||||
<img src="../assets/concepts/context/lost-in-middle.png" alt="Lost in the Middle paper illustration showing U-shaped attention curve across context positions" style="max-width: 420px; max-height: 55vh; width: auto; height: auto; display: block; border-radius: 8px; box-shadow: 0 4px 16px rgba(0,0,0,0.1); object-fit: contain;" />
|
||||
<img src="../assets/concepts/context/lost-in-the-middle.png" alt="Lost in the Middle paper illustration showing U-shaped attention curve across context positions" style="max-width: 420px; max-height: 55vh; width: auto; height: auto; display: block; border-radius: 8px; box-shadow: 0 4px 16px rgba(0,0,0,0.1); object-fit: contain;" />
|
||||
</div>
|
||||
|
||||
<!-- Right column: explanation -->
|
||||
@@ -2286,773 +2286,57 @@ todoapp/
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- ============================================================ -->
|
||||
<!-- TOPIC 2: SKILLS (Slides 15-20) -->
|
||||
<!-- ============================================================ -->
|
||||
|
||||
<!-- Slide 14: Skills Section Divider -->
|
||||
<div class="slide section-slide" data-slide="45" data-level="skills">
|
||||
<div class="section-number">Topic 2</div>
|
||||
<h1>🎓 Skills — What the Weather Reporter Knows</h1>
|
||||
<p class="section-desc">Skills are the specific things the reporter has been trained to do. Our reporter has two: <strong>fetch the data</strong>, and <strong>render it as a card</strong>.</p>
|
||||
</div>
|
||||
|
||||
<!-- Slide 15: The Training Manual -->
|
||||
<div class="slide" data-slide="46">
|
||||
<h1>The Training Manual</h1>
|
||||
<div class="analogy-box">
|
||||
<h4>Think of it like this</h4>
|
||||
<p>When someone joins your team, they have a role (agent), but they also go through specific training modules — how to use the CRM, how to write a proposal, how to run a standup. Each training module is a <strong>skill</strong>.</p>
|
||||
</div>
|
||||
<h3 style="margin-top: 32px;">The Weather Reporter's Two Skills</h3>
|
||||
<div class="use-cases">
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">🌡️</span>
|
||||
<div class="use-case-text">
|
||||
<strong>weather-fetcher</strong>
|
||||
<span>Knows exactly how to query Open-Meteo, extract the temperature value, and return it in the right format. This skill is <strong>preloaded into the agent's brain</strong> at startup.</span>
|
||||
</div>
|
||||
<!-- Slide 45: Workflow concept-intro -->
|
||||
<div class="slide section-slide" data-slide="45" data-level="workflow">
|
||||
<div class="slide-viewport-content">
|
||||
<h1>📘 Workflow</h1>
|
||||
<div style="text-align: center; margin: 12px auto 20px auto;">
|
||||
<span style="background: #fff3cd; color: #7a5c00; padding: 8px 20px; border-radius: 999px; font-size: 1rem; font-weight: 600; white-space: nowrap; box-shadow: 0 2px 8px rgba(122,92,0,0.18);">.claude/commands/</span>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">🖼️</span>
|
||||
<div class="use-case-text">
|
||||
<strong>weather-svg-creator</strong>
|
||||
<span>Takes a temperature value and renders a visual SVG weather card. Writes the output file to <code>orchestration-workflow/weather.svg</code>. This skill is invoked directly by the command.</span>
|
||||
</div>
|
||||
<p class="section-desc">Repeatable <strong>step-by-step recipes</strong> — the instruction manual that makes Claude run the same playbook every time.</p>
|
||||
<div style="display: flex; flex-wrap: wrap; gap: 10px; justify-content: center; margin-top: 28px;">
|
||||
<span style="background: #2e7d32; color: #fff; padding: 8px 20px; border-radius: 999px; font-size: 1rem; font-weight: 600; white-space: nowrap; box-shadow: 0 2px 8px rgba(46,125,50,0.35);">✅ Reproducible recipes</span>
|
||||
<span style="background: #2e7d32; color: #fff; padding: 8px 20px; border-radius: 999px; font-size: 1rem; font-weight: 600; white-space: nowrap; box-shadow: 0 2px 8px rgba(46,125,50,0.35);">🔁 Same steps, every run</span>
|
||||
<span style="background: #2e7d32; color: #fff; padding: 8px 20px; border-radius: 999px; font-size: 1rem; font-weight: 600; white-space: nowrap; box-shadow: 0 2px 8px rgba(46,125,50,0.35);">👥 Team-shareable via git</span>
|
||||
</div>
|
||||
</div>
|
||||
<p>Each skill has its own knowledge and methods. The reporter uses the right skill for the right step. <strong>Claude works the same way.</strong></p>
|
||||
</div>
|
||||
|
||||
<!-- Slide 16: When to Turn Something Into a Skill -->
|
||||
<div class="slide" data-slide="47">
|
||||
<h1>When to Turn Something Into a Skill</h1>
|
||||
<div class="info-box">
|
||||
<h4>Tip from Boris Cherny (creator of Claude Code) — Feb 1, 2026</h4>
|
||||
<p>"If you do something <strong>more than once a day</strong>, turn it into a skill."</p>
|
||||
</div>
|
||||
<p>The rule is simple: anything you're explaining to Claude over and over is a skill waiting to be written down. The weather reporter's skills are a perfect example:</p>
|
||||
<div class="use-cases">
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">🌡️</span>
|
||||
<div class="use-case-text">
|
||||
<strong>Fetch Dubai's temperature</strong>
|
||||
<span>Same API, same parameters, same format every time. That's <code>weather-fetcher</code>.</span>
|
||||
</div>
|
||||
<div class="pillar-footer">
|
||||
<div class="pillar-mini-card inactive" style="border-left: 3px solid #009688;">
|
||||
<span class="pmc-emoji">🧑‍💼</span>
|
||||
<span class="pmc-title">Agents</span>
|
||||
<span class="pmc-subtitle">— the specialists</span>
|
||||
<span class="pmc-body">A dedicated Claude worker — own context, own tools, own focus. Each runs in isolation.</span>
|
||||
<span class="pmc-badge" style="background: #e8f5e9; color: #2e7d32;">✅ fresh working memory per run</span>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">🖼️</span>
|
||||
<div class="use-case-text">
|
||||
<strong>Render the weather card</strong>
|
||||
<span>Same SVG template, same output path, same design every time. That's <code>weather-svg-creator</code>.</span>
|
||||
</div>
|
||||
<div class="pillar-mini-card inactive" style="border-left: 3px solid #9c27b0;">
|
||||
<span class="pmc-emoji">📝</span>
|
||||
<span class="pmc-title">CLAUDE.md</span>
|
||||
<span class="pmc-subtitle">— your memory</span>
|
||||
<span class="pmc-body">Knowledge <em>you</em> provide to the model. Read every session. Keep it under 200 lines.</span>
|
||||
<span class="pmc-badge" style="background: #fce4ec; color: #c62828;">⚠️ 200-line problem</span>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">📝</span>
|
||||
<div class="use-case-text">
|
||||
<strong>"Format my weekly update"</strong>
|
||||
<span>Same sections, same tone, same length every Monday. That's a skill waiting to be written.</span>
|
||||
</div>
|
||||
<div class="pillar-mini-card inactive" style="border-left: 3px solid #4caf50;">
|
||||
<span class="pmc-emoji">🎯</span>
|
||||
<span class="pmc-title">Skills</span>
|
||||
<span class="pmc-subtitle">— the know-how</span>
|
||||
<span class="pmc-body">What the specialist (or Claude) can actually do. Loaded on demand — only when needed.</span>
|
||||
<span class="pmc-badge" style="background: #e8f5e9; color: #2e7d32;">✅ progressive disclosure</span>
|
||||
</div>
|
||||
<div class="pillar-mini-card inactive" style="border-left: 3px solid #3f51b5;">
|
||||
<span class="pmc-emoji">💭</span>
|
||||
<span class="pmc-title">Context</span>
|
||||
<span class="pmc-subtitle">— the working memory</span>
|
||||
<span class="pmc-body">What Claude holds in his head now — fresh every new chat. Finite budget; compact at ~50%.</span>
|
||||
<span class="pmc-badge" style="background: #fce4ec; color: #c62828;">⚠️ dumb-zone problem</span>
|
||||
</div>
|
||||
<div class="pillar-mini-card" style="border-left: 4px solid #ef6c00;">
|
||||
<span class="pmc-emoji">📘</span>
|
||||
<span class="pmc-title">Workflows</span>
|
||||
<span class="pmc-subtitle">— the instruction manual</span>
|
||||
<span class="pmc-body">Repeatable step-by-step recipes — like an AC install guide. Deterministic, reproducible.</span>
|
||||
<span class="pmc-badge" style="background: #e8f5e9; color: #2e7d32;">✅ reproducible recipes</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="how-to-trigger">
|
||||
<h4>Takeaway</h4>
|
||||
<p>Skills are how <strong>repetition becomes reliability</strong>. Write it down once, and Claude does it the same way forever — or until you update the skill.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 17: Why Separate Agents and Skills? -->
|
||||
<div class="slide" data-slide="48">
|
||||
<h1>Why Separate Agents and Skills?</h1>
|
||||
<div class="two-col">
|
||||
<div>
|
||||
<h3>Agent = The Person</h3>
|
||||
<p>The <code>weather-agent</code> is the person with the job title <strong>"Weather Reporter"</strong>.</p>
|
||||
<p>It defines the <em>role</em> and <em>behavior</em>: use Open-Meteo, return temperature, track history in memory.</p>
|
||||
</div>
|
||||
<div>
|
||||
<h3>Skill = The Training</h3>
|
||||
<p>The <code>weather-fetcher</code> skill is the specific training on <strong>how</strong> to fetch weather data.</p>
|
||||
<p>It contains exact instructions:</p>
|
||||
<ul class="feature-list">
|
||||
<li>Call this specific API URL</li>
|
||||
<li>Extract <code>current.temperature_2m</code> from the response</li>
|
||||
<li>Return value + unit in this exact format</li>
|
||||
</ul>
|
||||
</div>
|
||||
</div>
|
||||
<div class="how-to-trigger" style="margin-top: 24px;">
|
||||
<h4>The Power</h4>
|
||||
<p><strong>One agent can have multiple skills</strong>, and <strong>one skill can be used by multiple agents</strong>. Skills are reusable building blocks. Agents are the people who use them.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 18: How to Create Your Own Skill -->
|
||||
<div class="slide" data-slide="49">
|
||||
<h1>How to Create Your Own Skill</h1>
|
||||
<p>Skills are plain markdown files. If you can write a recipe, you can write a skill.</p>
|
||||
<div class="how-to-trigger">
|
||||
<h4>The File</h4>
|
||||
<p>Create <code>.claude/skills/<your-skill-name>/SKILL.md</code> in your project. That's it.</p>
|
||||
</div>
|
||||
<div class="hiring-steps">
|
||||
<div class="hiring-step level-1">
|
||||
<span class="hiring-step-number">1</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Pick a repeated task</strong>
|
||||
<span>Something you explain to Claude more than once. "Fetch Dubai weather" is a perfect first skill.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-2">
|
||||
<span class="hiring-step-number">2</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Make a folder and a SKILL.md</strong>
|
||||
<span>Path: <code>.claude/skills/weather-fetcher/SKILL.md</code>. Or ask Claude: "Turn this into a skill."</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-3">
|
||||
<span class="hiring-step-number">3</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Write the recipe in plain English</strong>
|
||||
<span>Same words you'd use to explain it to a coworker. Skills are instructions, not code.</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="code-block"><span class="comment"># File: .claude/skills/weather-fetcher/SKILL.md</span>
|
||||
<span class="string">---</span>
|
||||
<span class="key">name</span>: <span class="string">weather-fetcher</span>
|
||||
<span class="key">description</span>: <span class="string">Fetch current temperature for Dubai from Open-Meteo API</span>
|
||||
<span class="key">user-invocable</span>: <span class="string">false</span>
|
||||
<span class="string">---</span>
|
||||
|
||||
Fetch the current temperature for Dubai, UAE.
|
||||
|
||||
1. Call the Open-Meteo API:
|
||||
<span class="cmd">https://api.open-meteo.com/v1/forecast</span>
|
||||
<span class="cmd">?latitude=25.2048&longitude=55.2708¤t=temperature_2m</span>
|
||||
|
||||
2. Extract <span class="key">current.temperature_2m</span> from the response
|
||||
|
||||
3. Return: <span class="string">"The temperature in Dubai is {value}{unit}"</span></div>
|
||||
<div class="trigger-box">
|
||||
<h4>Takeaway</h4>
|
||||
<p>The skill contains <strong>instructions, not code</strong>. It tells Claude <em>how</em> to do something using its existing tools.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 19: Skill Config Fields -->
|
||||
<div class="slide" data-slide="50">
|
||||
<h1>Skill Config Fields</h1>
|
||||
<p>The small config block at the top of a SKILL.md (the "frontmatter") controls how the skill behaves:</p>
|
||||
<div style="margin: 24px 0;">
|
||||
<div class="field-row">
|
||||
<span class="field-name">name</span>
|
||||
<span class="field-desc">Display name and <code>/slash-command</code>. Defaults to directory name</span>
|
||||
</div>
|
||||
<div class="field-row">
|
||||
<span class="field-name">description <span class="field-recommended">Recommended</span></span>
|
||||
<span class="field-desc">What the skill does — shown in autocomplete, used for auto-discovery</span>
|
||||
</div>
|
||||
<div class="field-row">
|
||||
<span class="field-name">user-invocable</span>
|
||||
<span class="field-desc">Set <code>false</code> to hide from <code>/</code> menu — <strong>weather-fetcher uses this</strong> because it's agent-only</span>
|
||||
</div>
|
||||
<div class="field-row">
|
||||
<span class="field-name">allowed-tools</span>
|
||||
<span class="field-desc">Tools allowed without permission prompts when skill is active</span>
|
||||
</div>
|
||||
<div class="field-row">
|
||||
<span class="field-name">model</span>
|
||||
<span class="field-desc">Model to use: <code>haiku</code>, <code>sonnet</code>, or <code>opus</code></span>
|
||||
</div>
|
||||
<div class="field-row">
|
||||
<span class="field-name">context</span>
|
||||
<span class="field-desc">Set to <code>fork</code> to run in isolated subagent context</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="info-box">
|
||||
<h4>Two Ways to Use Skills</h4>
|
||||
<p><strong>User-invocable</strong>: appears in <code>/</code> menu, user runs it directly (like <code>weather-svg-creator</code>). <strong>Agent-preloaded</strong>: set <code>user-invocable: false</code>, then list it in an agent's <code>skills:</code> field — it becomes domain knowledge injected into that agent (like <code>weather-fetcher</code>).</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- ============================================================ -->
|
||||
<!-- TOPIC 3: CONTEXT (Slides 21-25) -->
|
||||
<!-- ============================================================ -->
|
||||
|
||||
<!-- Slide 20: Context Section Divider -->
|
||||
<div class="slide section-slide" data-slide="51" data-level="context">
|
||||
<div class="section-number">Topic 3</div>
|
||||
<h1>🧠 Context — The Reporter's Brain</h1>
|
||||
<p class="section-desc">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.</p>
|
||||
</div>
|
||||
|
||||
<!-- Slide 21: Claude's Brain -->
|
||||
<div class="slide" data-slide="52">
|
||||
<h1>Claude's Brain</h1>
|
||||
<div class="analogy-box">
|
||||
<h4>Think of it like this</h4>
|
||||
<p>Imagine Claude has a <strong>brain</strong> 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.</p>
|
||||
</div>
|
||||
<img src="../assets/concepts/context-window.jpeg" alt="Context window diagram showing how the 1M token limit is divided across system prompt, tools, files, and conversation" style="width: 100%; max-width: 800px; margin: 24px auto; display: block; border-radius: 8px;" />
|
||||
<p>When the <code>weather-agent</code> is dispatched, it gets its own fresh brain — and <code>weather-fetcher</code> is pinned into it at startup via the <code>skills:</code> field. Everything you give Claude in a conversation lands in its brain:</p>
|
||||
<div class="use-cases">
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">💬</span>
|
||||
<div class="use-case-text">
|
||||
<strong>Your messages</strong>
|
||||
<span>Every question, clarification, and follow-up</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">📄</span>
|
||||
<div class="use-case-text">
|
||||
<strong>Every file Claude opens</strong>
|
||||
<span>Read a 500-page PDF? All of it is in the brain now.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">🔧</span>
|
||||
<div class="use-case-text">
|
||||
<strong>Every tool result</strong>
|
||||
<span>Web searches, command output, screenshots — all held in memory</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="warning-box">
|
||||
<h4>Two Things Every Non-Engineer Should Know</h4>
|
||||
<p><strong>1. The brain is finite.</strong> It can hold about 1 million tokens — roughly 750,000 words, or a short novel. Big, but not infinite. <strong>2. The brain empties at the end of every chat.</strong> When you start a new conversation, Claude remembers <em>nothing</em> from the last one unless you tell it again.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 22: What Loads at Session Start -->
|
||||
<div class="slide" data-slide="53">
|
||||
<h1>What Loads at Session Start</h1>
|
||||
<p>The moment you open Claude Code, certain things land in Claude's brain <strong>before you've typed a word</strong>. The rest waits in the wings — only loaded when you actually need it. This is called <strong>progressive disclosure</strong>.</p>
|
||||
<img src="../assets/concepts/context.jpg" alt="Diagram showing what loads into Claude's context window at session start" style="width: 100%; max-width: 800px; margin: 24px auto; display: block; border-radius: 8px;" />
|
||||
<div class="use-cases" style="margin-top: 24px;">
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">📋</span>
|
||||
<div class="use-case-text">
|
||||
<strong>CLAUDE.md — full content</strong>
|
||||
<span>Standing instructions, every line. Always pinned in the brain at startup.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">🎓</span>
|
||||
<div class="use-case-text">
|
||||
<strong>Skills — descriptions only</strong>
|
||||
<span>Claude knows which skills exist and what each does. At startup, Claude knows <em>about</em> <code>weather-fetcher</code>; the full content loads when the agent runs.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">👤</span>
|
||||
<div class="use-case-text">
|
||||
<strong>Agents — descriptions only</strong>
|
||||
<span>The roster of specialists. The weather-agent's full system prompt loads when it's dispatched.</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="trigger-box">
|
||||
<h4>The One-Liner</h4>
|
||||
<p>Only <strong>descriptions</strong> of skills and agents are loaded at startup — the rest is fetched on-demand. That's <strong>progressive disclosure</strong>. It keeps the brain light.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 23: Keep the Brain Clear -->
|
||||
<div class="slide" data-slide="54">
|
||||
<h1>Keep the Brain Clear</h1>
|
||||
<p>The more stuff crammed into Claude's brain, the harder it is to focus on what matters. This is called <strong>context rot</strong> — performance drops as the brain gets crowded.</p>
|
||||
<div class="info-box">
|
||||
<h4>Tip from Thariq (Anthropic) — Apr 16, 2026</h4>
|
||||
<p>"Every turn is a branching point." After Claude finishes a response, you don't have to just keep typing. You have <strong>five options</strong> — and four of them exist to keep the brain clear.</p>
|
||||
</div>
|
||||
<table>
|
||||
<thead>
|
||||
<tr>
|
||||
<th>Option</th>
|
||||
<th>When to use it</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td><strong>Continue</strong></td>
|
||||
<td>Same task, everything in the brain still matters</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td><strong>Rewind</strong> (tap Esc twice)</td>
|
||||
<td>Claude went down a wrong path — jump back and try again</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td><strong>/compact</strong></td>
|
||||
<td>The brain is bloated — summarize and keep going</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td><strong>/clear</strong> (fresh chat)</td>
|
||||
<td>Starting a truly new task — wipe the brain clean</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td><strong>Subagent</strong></td>
|
||||
<td>Next step will make a mess — send a helper to do it in <em>their own</em> brain</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<div class="how-to-trigger">
|
||||
<h4>Rule of Thumb</h4>
|
||||
<p>When you start a <strong>new task</strong>, start a <strong>new chat</strong>. Don't let yesterday's thinking clutter today's brain.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 24: How to Manage Your Context -->
|
||||
<div class="slide" data-slide="55">
|
||||
<h1>How to Manage Your Context</h1>
|
||||
<p>You can't <em>create</em> the context — it's just there, the moment you open a chat. But you can <strong>see</strong> how full it is, <strong>trim</strong> it down, or <strong>wipe</strong> it clean. Three commands give you full control.</p>
|
||||
<div class="how-to-trigger">
|
||||
<h4>The Commands</h4>
|
||||
<p><code>/context</code> shows you how full the brain is. <code>/compact</code> summarizes it. <code>/clear</code> wipes it.</p>
|
||||
</div>
|
||||
<div class="hiring-steps">
|
||||
<div class="hiring-step level-1">
|
||||
<span class="hiring-step-number">1</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Check how full the brain is — <code>/context</code></strong>
|
||||
<span>Shows a breakdown of how many tokens are used and what's taking up space (system prompt, tools, files, conversation).</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-2">
|
||||
<span class="hiring-step-number">2</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Trim or reset — <code>/compact</code> or <code>/clear</code></strong>
|
||||
<span><code>/compact</code> keeps a short summary so you can continue. <code>/clear</code> wipes everything for a truly fresh start.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-3">
|
||||
<span class="hiring-step-number">3</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Start the next task fresh</strong>
|
||||
<span>The brain is light again — re-state the goal, and Claude works without yesterday's clutter.</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="code-block"><span class="comment"># The three context commands</span>
|
||||
<span class="cmd">/context</span> <span class="comment"># Check how full the brain is right now</span>
|
||||
<span class="cmd">/compact</span> <span class="comment"># Summarize what's there and keep going on top</span>
|
||||
<span class="cmd">/clear</span> <span class="comment"># Wipe the brain — fresh chat, nothing remembered</span></div>
|
||||
<div class="trigger-box">
|
||||
<h4>Takeaway</h4>
|
||||
<p>Check first with <code>/context</code>. Trim with <code>/compact</code>. Reset with <code>/clear</code> when starting a new task.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- ============================================================ -->
|
||||
<!-- TOPIC 4: CLAUDE.md (Slides 26-31) -->
|
||||
<!-- ============================================================ -->
|
||||
|
||||
<!-- Slide 25: CLAUDE.md Section Divider -->
|
||||
<div class="slide section-slide" data-slide="56" data-level="claude-md">
|
||||
<div class="section-number">Topic 4</div>
|
||||
<h1>📋 CLAUDE.md — The Reporter's Pocket Rulebook</h1>
|
||||
<p class="section-desc">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.</p>
|
||||
</div>
|
||||
|
||||
<!-- Slide 26: The Employee Handbook -->
|
||||
<div class="slide" data-slide="57">
|
||||
<h1>The Employee Handbook</h1>
|
||||
<div class="analogy-box">
|
||||
<h4>Think of it like this</h4>
|
||||
<p>Imagine hiring a brilliant new assistant every single morning — someone who forgets everything at 5pm. You'd get tired of re-explaining "we say <em>clients</em>, not <em>customers</em>" and "always CC my manager" over and over. So you pin those rules into their brain before the day starts. That's <code>CLAUDE.md</code>: a short list of standing instructions Claude reads first, every time.</p>
|
||||
</div>
|
||||
<p>For the weather reporter, CLAUDE.md might contain: "Always report in Celsius unless asked. Always cite Open-Meteo as the source." Here's a non-technical example:</p>
|
||||
<div class="code-block"><span class="comment"># CLAUDE.md</span>
|
||||
|
||||
<span class="comment">## About this project</span>
|
||||
This is the Q2 marketing campaign brief — targeting small business owners.
|
||||
|
||||
<span class="comment">## How I want you to talk to me</span>
|
||||
- Always explain technical terms in plain language
|
||||
- Before you run any command, tell me what it does in one sentence
|
||||
- If I ask for an opinion, give me one — don't hedge
|
||||
|
||||
<span class="comment">## House rules</span>
|
||||
- We say "clients" not "customers"
|
||||
- Never edit files in the /archive folder
|
||||
- Keep emails under 150 words</div>
|
||||
<div class="how-to-trigger">
|
||||
<h4>You Don't Have to Write It From Scratch</h4>
|
||||
<p>Claude can draft your first CLAUDE.md for you. The next slide shows how.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 27: How to Create CLAUDE.md -->
|
||||
<div class="slide" data-slide="58">
|
||||
<h1>How to Create Your CLAUDE.md</h1>
|
||||
<p>You don't need to write CLAUDE.md by hand. Claude can look at your project and draft one for you.</p>
|
||||
<div class="how-to-trigger">
|
||||
<h4>The Command</h4>
|
||||
<p>Type <code>/init</code> inside Claude Code, in your project folder. Claude does the rest.</p>
|
||||
</div>
|
||||
<div class="hiring-steps">
|
||||
<div class="hiring-step level-1">
|
||||
<span class="hiring-step-number">1</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Type <code>/init</code> in Claude Code</strong>
|
||||
<span>Make sure you're in your project folder. No other setup needed.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-2">
|
||||
<span class="hiring-step-number">2</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Claude reads your project</strong>
|
||||
<span>It looks at your files, folder names, and README to figure out what the project is about.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-3">
|
||||
<span class="hiring-step-number">3</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Claude drafts a starter CLAUDE.md</strong>
|
||||
<span>Saved to the root of your project. Open it and edit like any other document.</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="analogy-box">
|
||||
<h4>Think of it like this</h4>
|
||||
<p>Running <code>/init</code> is like <strong>asking the new hire to read everything on the shared drive before you meet them</strong> — they walk in already knowing the project's shape, and they write up what they learned as a handbook for future Claude.</p>
|
||||
</div>
|
||||
<div class="code-block"><span class="comment"># The 60-second workflow</span>
|
||||
<span class="cmd">/init</span> <span class="comment"># Claude drafts the file</span>
|
||||
<span class="claude-file">CLAUDE.md</span> appears <span class="comment"># In your project root</span>
|
||||
open, edit, save <span class="comment"># Tweak it like any doc</span></div>
|
||||
<div class="trigger-box">
|
||||
<h4>Takeaway</h4>
|
||||
<p>You'll keep improving it over time. But <code>/init</code> gets you to a working draft in under a minute.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 28: Grow CLAUDE.md With Every Mistake -->
|
||||
<div class="slide" data-slide="59">
|
||||
<h1>Grow CLAUDE.md With Every Mistake</h1>
|
||||
<div class="info-box">
|
||||
<h4>Tip from Boris Cherny (creator of Claude Code) — Feb 1, 2026</h4>
|
||||
<p>"After every correction, end with: <em>Update your CLAUDE.md so you don't make that mistake again.</em> Claude is eerily good at writing rules for itself."</p>
|
||||
</div>
|
||||
<p>The CLAUDE.md isn't something you write once and forget. It's a <strong>living memory</strong> of every lesson you've ever had to teach Claude. The workflow:</p>
|
||||
<div class="hiring-steps">
|
||||
<div class="hiring-step level-1">
|
||||
<span class="hiring-step-number">1</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Claude makes a mistake</strong>
|
||||
<span>It uses the wrong tone, touches a file you didn't want touched, skips a step</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-2">
|
||||
<span class="hiring-step-number">2</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>You correct it</strong>
|
||||
<span>"Actually, we never write in all caps. And always confirm before sending."</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-3">
|
||||
<span class="hiring-step-number">3</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>You say: "update your CLAUDE.md so this doesn't happen again"</strong>
|
||||
<span>Claude writes the new rule for itself. Next conversation, it won't repeat the mistake.</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="warning-box">
|
||||
<h4>One Rule</h4>
|
||||
<p>Keep it <strong>short</strong>. Under 200 lines. Long CLAUDE.md files get ignored — same as a 50-page employee handbook nobody reads.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 29: What Goes in CLAUDE.md -->
|
||||
<div class="slide" data-slide="60">
|
||||
<h1>What Goes in CLAUDE.md</h1>
|
||||
<div class="code-block"><span class="comment"># CLAUDE.md</span>
|
||||
|
||||
<span class="comment">## Project Overview</span>
|
||||
This is a TodoApp with a FastAPI backend and React frontend.
|
||||
|
||||
<span class="comment">## Key Commands</span>
|
||||
- <span class="cmd">npm run dev</span> <span class="comment"># Start frontend</span>
|
||||
- <span class="cmd">uvicorn main:app</span> <span class="comment"># Start backend</span>
|
||||
- <span class="cmd">pytest</span> <span class="comment"># Run tests</span>
|
||||
|
||||
<span class="comment">## Architecture</span>
|
||||
- Routes follow the pattern in <span class="string">routes/todos.py</span>
|
||||
- Frontend components use Tailwind CSS
|
||||
- Tests mirror the source tree: <span class="string">tests/test_*.py</span>
|
||||
|
||||
<span class="comment">## Rules</span>
|
||||
- Always write tests for new endpoints
|
||||
- Use the existing Sidebar component for navigation
|
||||
- Keep CLAUDE.md under 200 lines</div>
|
||||
<div class="warning-box">
|
||||
<h4>Keep It Short</h4>
|
||||
<p>Aim for under 200 lines. Files over 200 lines consume more context and may reduce adherence. Put detailed instructions in <code>.claude/rules/</code> instead.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 30: How CLAUDE.md Loads -->
|
||||
<div class="slide" data-slide="61">
|
||||
<h1>How CLAUDE.md Loads</h1>
|
||||
<p>Claude Code uses two mechanisms to find CLAUDE.md files:</p>
|
||||
<div class="two-col">
|
||||
<div class="col-card" style="border-left: 4px solid #2196f3;">
|
||||
<h4>Ancestor Loading (UP)</h4>
|
||||
<p>At startup, Claude walks <strong>upward</strong> from your current directory and loads every CLAUDE.md it finds.</p>
|
||||
<p style="margin-top: 12px;"><strong>Always loaded immediately.</strong></p>
|
||||
</div>
|
||||
<div class="col-card" style="border-left: 4px solid #ff9800;">
|
||||
<h4>Descendant Loading (DOWN)</h4>
|
||||
<p>CLAUDE.md files in subdirectories below you are loaded <strong>lazily</strong> — only when Claude reads files in those folders.</p>
|
||||
<p style="margin-top: 12px;"><strong>Loaded on demand.</strong></p>
|
||||
</div>
|
||||
</div>
|
||||
<div class="code-block"><span class="comment"># Example: running `claude` from /myapp</span>
|
||||
/myapp/
|
||||
<span class="claude-file">CLAUDE.md</span> <span class="comment"># Loaded at startup (current dir)</span>
|
||||
frontend/
|
||||
<span class="key">CLAUDE.md</span> <span class="comment"># Loaded when you touch frontend/ files</span>
|
||||
backend/
|
||||
<span class="key">CLAUDE.md</span> <span class="comment"># Loaded when you touch backend/ files</span></div>
|
||||
<div class="info-box">
|
||||
<h4>Rules Directory</h4>
|
||||
<p>For topic-specific instructions, use <code>.claude/rules/*.md</code>. Each rule file can target specific file patterns using <code># Glob: pattern</code> at the top — they only load when Claude works with matching files.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- ============================================================ -->
|
||||
<!-- TOPIC 5: COMMANDS (Slides 32-34) -->
|
||||
<!-- ============================================================ -->
|
||||
|
||||
<!-- Slide 31: Commands Section Divider -->
|
||||
<div class="slide section-slide" data-slide="62" data-level="commands">
|
||||
<div class="section-number">Topic 5</div>
|
||||
<h1>⚡ Commands — The Trigger</h1>
|
||||
<p class="section-desc">One word kicks off the whole chain. <code>/weather-orchestrator</code> → agent → skill → SVG card. Commands are the entry point into any workflow.</p>
|
||||
</div>
|
||||
|
||||
<!-- Slide 32: Commands — The Entry Point -->
|
||||
<div class="slide" data-slide="63">
|
||||
<h1>Commands — The Entry Point</h1>
|
||||
<div class="analogy-box">
|
||||
<h4>Think of it like this</h4>
|
||||
<p>If agents are employees and skills are their training, a <strong>command</strong> is the manager's playbook — "when someone asks for a weather report, first ask them Celsius or Fahrenheit, then dispatch the weather team, then create the visual."</p>
|
||||
</div>
|
||||
<p>You've already seen built-in commands in this presentation:</p>
|
||||
<div class="use-cases">
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">📋</span>
|
||||
<div class="use-case-text">
|
||||
<strong><code>/init</code></strong>
|
||||
<span>Drafts a CLAUDE.md for your project (Topic 4)</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">👤</span>
|
||||
<div class="use-case-text">
|
||||
<strong><code>/agents</code></strong>
|
||||
<span>Opens the agent creator menu (Topic 1)</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">🧠</span>
|
||||
<div class="use-case-text">
|
||||
<strong><code>/context</code></strong>
|
||||
<span>Shows how full Claude's brain is right now (Topic 3)</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="use-case-item">
|
||||
<span class="use-case-icon">🧹</span>
|
||||
<div class="use-case-text">
|
||||
<strong><code>/clear</code> & <code>/compact</code></strong>
|
||||
<span>Reset or summarize the brain (Topic 3)</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="how-to-trigger">
|
||||
<h4>The Magic</h4>
|
||||
<p>You can write your <strong>own</strong> commands the same way. The next slide shows how.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 33: How to Create Your Own Command -->
|
||||
<div class="slide" data-slide="64">
|
||||
<h1>How to Create Your Own Command</h1>
|
||||
<p>Commands are markdown files too. If you can write a recipe, you can write a command.</p>
|
||||
<div class="how-to-trigger">
|
||||
<h4>The File</h4>
|
||||
<p>Create <code>.claude/commands/<your-command-name>.md</code>. As soon as it's saved, <code>/<your-command-name></code> appears in Claude Code's slash menu.</p>
|
||||
</div>
|
||||
<div class="hiring-steps">
|
||||
<div class="hiring-step level-1">
|
||||
<span class="hiring-step-number">1</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Pick a workflow you trigger often</strong>
|
||||
<span>"Generate this week's status report." Anything you'd want one keystroke for.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-2">
|
||||
<span class="hiring-step-number">2</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Create the file</strong>
|
||||
<span>Path: <code>.claude/commands/weather-orchestrator.md</code>. The filename becomes the command name.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-3">
|
||||
<span class="hiring-step-number">3</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Write the playbook in plain English</strong>
|
||||
<span>Step-by-step what Claude should do — including which agents to dispatch and which skills to invoke.</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="analogy-box">
|
||||
<h4>Think of it like this</h4>
|
||||
<p>A command is like <strong>a button on a coffee machine</strong>. One press, the machine grinds, brews, foams, pours. You don't think about the steps — you just want espresso. Slash commands are buttons you've labelled yourself.</p>
|
||||
</div>
|
||||
<div class="code-block"><span class="comment"># File: .claude/commands/weather-orchestrator.md</span>
|
||||
<span class="string">---</span>
|
||||
<span class="key">description</span>: <span class="string">Fetch weather data for Dubai and create an SVG weather card</span>
|
||||
<span class="key">model</span>: <span class="string">haiku</span>
|
||||
<span class="string">---</span>
|
||||
|
||||
1. Ask the user: Celsius or Fahrenheit?
|
||||
2. Use the <span class="key">weather-agent</span> to fetch Dubai's temperature
|
||||
3. Use the <span class="key">weather-svg-creator</span> skill to create an SVG card
|
||||
4. Show the user the result</div>
|
||||
<div class="trigger-box">
|
||||
<h4>Takeaway</h4>
|
||||
<p>Commands are how you turn a multi-step prompt into a single keystroke that anyone on your team can run.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- ============================================================ -->
|
||||
<!-- TOPIC 6: WORKFLOW (Slides 35-38) -->
|
||||
<!-- ============================================================ -->
|
||||
|
||||
<!-- Slide 34: Workflow Sub-Section (Putting It All Together) -->
|
||||
<div class="slide section-slide" data-slide="65" data-level="workflow">
|
||||
<div class="section-number">Putting It All Together</div>
|
||||
<h1>🎼 Workflow — All Five Pieces Together</h1>
|
||||
<p class="section-desc">Watch the weather reporter example run from one keystroke to SVG card output. Five concepts, one orchestrated flow.</p>
|
||||
</div>
|
||||
|
||||
<!-- Slide 35: Command → Agent → Skill -->
|
||||
<div class="slide" data-slide="66">
|
||||
<h1>Command → Agent → Skill</h1>
|
||||
<p>This is the <strong>core architecture pattern</strong> of Claude Code workflows — demonstrated in this very repo by the weather example:</p>
|
||||
<div class="code-block"><span class="comment">The Orchestration Flow</span>
|
||||
|
||||
<span class="key">/weather-orchestrator</span> <span class="comment">COMMAND (entry point)</span>
|
||||
|
|
||||
Step 1: Ask user C/F?
|
||||
|
|
||||
Step 2: <span class="cmd">Agent tool</span>
|
||||
|
|
||||
v
|
||||
<span class="key">weather-agent</span> <span class="comment">AGENT (specialist)</span>
|
||||
skill: <span class="string">weather-fetcher</span> <span class="comment">SKILL (preloaded training)</span>
|
||||
tools: WebFetch, Read
|
||||
|
|
||||
Returns: temp + unit
|
||||
|
|
||||
Step 3: <span class="cmd">Skill tool</span>
|
||||
|
|
||||
v
|
||||
<span class="key">weather-svg-creator</span> <span class="comment">SKILL (invoked directly)</span>
|
||||
|
|
||||
Creates: weather.svg + output.md</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 36: Two Ways Skills Are Used -->
|
||||
<div class="slide" data-slide="67">
|
||||
<h1>Two Ways Skills Are Used</h1>
|
||||
<p>The weather workflow demonstrates both skill patterns in a single flow:</p>
|
||||
<div class="two-col">
|
||||
<div class="col-card" style="border-left: 4px solid #9c27b0;">
|
||||
<h4>Pattern 1: Agent Skill (Preloaded)</h4>
|
||||
<p><code>weather-fetcher</code> is listed in the agent's <code>skills:</code> field.</p>
|
||||
<p style="margin-top: 12px;">Its full content is <strong>injected into the agent's context</strong> at startup. The agent reads it like a reference manual.</p>
|
||||
<div class="code-block" style="font-size: 0.8rem;"><span class="comment"># In weather-agent.md</span>
|
||||
<span class="key">skills</span>:
|
||||
- <span class="string">weather-fetcher</span></div>
|
||||
</div>
|
||||
<div class="col-card" style="border-left: 4px solid #2196f3;">
|
||||
<h4>Pattern 2: Direct Invocation</h4>
|
||||
<p><code>weather-svg-creator</code> is called directly by the command via the Skill tool.</p>
|
||||
<p style="margin-top: 12px;">It runs <strong>independently</strong> in the command's context, not inside any agent.</p>
|
||||
<div class="code-block" style="font-size: 0.8rem;"><span class="comment"># In the command prompt</span>
|
||||
<span class="cmd">Skill</span>(<span class="key">skill</span>: <span class="string">"weather-svg-creator"</span>)</div>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- Slide 37: How to Wire Your Own Workflow -->
|
||||
<div class="slide" data-slide="68">
|
||||
<h1>How to Wire Your Own Workflow</h1>
|
||||
<p>A workflow isn't a separate file type. It <em>emerges</em> when one command calls agents and skills in sequence.</p>
|
||||
<div class="how-to-trigger">
|
||||
<h4>The Recipe</h4>
|
||||
<p>One <strong>command</strong> at the top. One <strong>agent</strong> with a <strong>skill</strong> preloaded for fetching. One <strong>skill</strong> at the end for output. Wire them in the command file.</p>
|
||||
</div>
|
||||
<div class="hiring-steps">
|
||||
<div class="hiring-step level-1">
|
||||
<span class="hiring-step-number">1</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Write the command (entry point)</strong>
|
||||
<span><code>.claude/commands/weather-orchestrator.md</code> — the playbook the user types.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-2">
|
||||
<span class="hiring-step-number">2</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Write the agent and its preloaded skill</strong>
|
||||
<span><code>.claude/agents/weather-agent.md</code> with <code>skills: [weather-fetcher]</code> in its config — the specialist + their training.</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="hiring-step level-3">
|
||||
<span class="hiring-step-number">3</span>
|
||||
<div class="hiring-step-content">
|
||||
<strong>Write the output skill</strong>
|
||||
<span><code>.claude/skills/weather-svg-creator/SKILL.md</code> — invoked directly by the command after the agent returns.</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<div class="analogy-box">
|
||||
<h4>Think of it like this</h4>
|
||||
<p>The workflow is the <strong>relay race</strong>. The command is the starter pistol. The agent is runner one (fetches the data, hands the baton back). The skill is runner two (turns the data into a visual). You don't run the race — you just point and say "go."</p>
|
||||
</div>
|
||||
<div class="code-block"><span class="comment"># The weather workflow — three files, one keystroke</span>
|
||||
<span class="claude-file">.claude/commands/weather-orchestrator.md</span> <span class="comment"># Entry point</span>
|
||||
<span class="claude-file">.claude/agents/weather-agent.md</span> <span class="comment"># Specialist + skill</span>
|
||||
<span class="claude-file">.claude/skills/weather-fetcher/SKILL.md</span> <span class="comment"># Preloaded into agent</span>
|
||||
<span class="claude-file">.claude/skills/weather-svg-creator/SKILL.md</span> <span class="comment"># Invoked by command</span>
|
||||
|
||||
<span class="comment"># User types one thing:</span>
|
||||
<span class="cmd">/weather-orchestrator</span></div>
|
||||
<div class="trigger-box">
|
||||
<h4>Takeaway</h4>
|
||||
<p>Every workflow you'll ever build follows this same shape: <strong>Command at the top, Agent + Skill in the middle, Skill at the end</strong>. Once you've wired one, the rest are variations on the theme.</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- ============================================================ -->
|
||||
<!-- SLIDE 40: Closing -->
|
||||
<!-- ============================================================ -->
|
||||
<div class="slide title-slide" data-slide="69">
|
||||
<h1>Journey So Far</h1>
|
||||
<p class="subtitle">Five concepts, one running example</p>
|
||||
<p style="margin-top: 20px; font-size: 1.1rem; color: #666; max-width: 640px;">From meeting the weather reporter to wiring the full <strong>Command → Agent → Skill</strong> chain. The same five pieces compose every workflow you'll ever build.</p>
|
||||
<div style="margin-top: 40px; text-align: left; max-width: 500px;">
|
||||
<p style="font-size: 1rem; color: #888; margin-bottom: 8px;">👤 Agents — the specialist with a job title</p>
|
||||
<p style="font-size: 1rem; color: #888; margin-bottom: 8px;">🎓 Skills — the training they carry</p>
|
||||
<p style="font-size: 1rem; color: #888; margin-bottom: 8px;">🧠 Context — the brain that resets each session</p>
|
||||
<p style="font-size: 1rem; color: #888; margin-bottom: 8px;">📋 CLAUDE.md — the pocket rulebook that survives resets</p>
|
||||
<p style="font-size: 1rem; color: #888; margin-bottom: 8px;">⚡ Commands & 🎼 Workflow — the trigger that wires it all</p>
|
||||
</div>
|
||||
<p style="margin-top: 40px; font-size: 0.95rem; color: #aaa;">github.com/shanraisshan/claude-code-best-practice</p>
|
||||
</div>
|
||||
|
||||
<!-- Corner mascots (global, appear on every slide) -->
|
||||
|
||||
|
Before Width: | Height: | Size: 34 KiB After Width: | Height: | Size: 34 KiB |
|
Before Width: | Height: | Size: 564 KiB After Width: | Height: | Size: 564 KiB |
Reference in New Issue
Block a user