myslide

name	myslide
description	Create professional AWS-themed PowerPoint presentations with two theme options: dark (reInvent 2023/2025 with gradient backgrounds) and light (L100/field enablement with white backgrounds and soft gradient accents). Uses AWS brand colors, rich visual elements (SVG diagrams, architecture diagrams, icons), and cross-skill integration (svg-diagram). Supports creating from scratch, editing specific slides conversationally, and embedding SVG visualizations for maximum design impact. Trigger: "myslide", "make slides", "AWS presentation", "create pptx", "slide deck", "AWS slides", "프레젠테이션", "슬라이드 만들어", "발표자료", "L100", "L200", "training deck", "light theme", "밝은 테마", "교육 자료"
license	MIT License
metadata	{"skill-author":"Jesam Kim","version":"1.0.0"}
allowed-tools	["Read","Write","Edit","Bash","Grep","Glob","Agent","AskUserQuestion"]

MySlide - AWS-Themed Presentation Generator

Create visually compelling presentations that follow AWS design systems. Supports two themes: dark (reInvent 2023/2025) and light (L100/field enablement). Every slide should look like it was crafted by the AWS brand team.

Theme Selection

Theme	When to Use	Reference
Dark (reInvent 2023/2025)	reInvent, Summit keynotes, tech demos	aws-theme.md
Light (L100/Field Enablement)	L100/L200 training, customer-facing, internal workshops, sales enablement	light-theme.md

Auto-detect: If the user says "L100", "training deck", "white background", "밝은 테마", "교육 자료", "customer-facing" → use Light theme. Otherwise default to Dark theme.

Quick Start

1. Choose theme: Dark (default) or Light — read the corresponding theme reference
2. Read references/aws-theme.md for dark theme OR references/light-theme.md for light theme
3. Read references/slide-patterns.md for layout templates
4. Read references/pptxgenjs.md for PptxGenJS creation guide
5. Read references/editing.md for editing existing PPTX files
6. Read references/animations.md for animation primitives
7. Use scripts/create_aws_slide.py to generate background/SVG assets
8. Official AWS service icons are in icons/ (304 icons, including 56 Bedrock AgentCore variants)

All references and scripts are self-contained within this skill directory. No external skill dependencies required.

Cross-Skill Integration (Design Enhancement)

For richer visual output, leverage these companion skills when available:

• svg-diagram: Generate pixel-perfect SVG diagrams, architecture visuals, and flowcharts with anti-overlap rules. Use for any slide needing diagrams beyond basic arrow connections.

Directory Convention — All generated assets go under artifacts/ (sibling to the skill directory), never /tmp:

ASSETS_DIR  = {workspace}/artifacts/myslide-assets/   # backgrounds, SVG/PNG assets, generated images
PARTS_DIR   = {workspace}/artifacts/myslide-parts/    # individual slide JS snippets
QA_DIR      = {workspace}/artifacts/myslide-qa/       # QA render outputs (PDF, JPEG)

{workspace} is the project root (the directory containing application/).

Skill root (this repo)

The system prompt exposes WORKING_DIR as the application/ directory. This skill lives only here:

Path	Meaning
{WORKING_DIR}/skills/myslide	Root of this skill (SKILL.md, scripts/, icons/, references/)

• Use {WORKING_DIR}/skills/myslide for every script path, read_file, and cd.
• Optional helper skills (sd35l, nova2-omni, kiro, etc.) must be resolved under {WORKING_DIR}/skills/<name> (same pattern as myslide).

# Example: locate sd35l under application/skills
SD35L_SCRIPT=$(find "${WORKING_DIR}/skills" -path "*/sd35l/scripts/generate_image.py" 2>/dev/null | head -1)

Default Presenter

When generating title/thank-you slides, use these defaults unless the user specifies otherwise:

• Korean: 발표자
• English: Name
• Title: Solutions Architect
• Company: Amazon Web Services
• Email: [e-mail]

Workflow

A. Creating a New Presentation

1. Gather requirements: Ask the user for topic, key messages, and target audience

2. Write a design spec (when applicable — see gate below): Produce a markdown spec that lists every slide's layout, key message, and visual intent. Save to design-specs/<deck-name>.md. See references/design-spec-template.md for the table format and approval flow.

   Spec gate — when to require a spec before any PPTX work:

   Scope	Spec?	HTML preview?
   1-2 slides, single edit	No	No
   3-7 slide deck	Yes (markdown only)	Optional
   8+ slide deck	Yes	Yes
   User says "design first" / "디자인 먼저" / "plan first"	Always	Always

   The reason: structural rework (wrong slide order, monotonous layout, wrong theme) is the most expensive kind to fix once PptxGenJS code exists. A spec catches it in seconds. For a quick one-pager the gate adds friction without the savings, so skip it.

   Approval is mandatory for decks where the gate applies. After writing the spec (and rendering the preview if applicable), wait for the user to say "go" / "승인" / "OK" / "진행" before moving to step 3. If they ask for changes, edit the spec, re-render, ask again — don't start building from a partially-approved spec.

   HTML preview (8+ slides):

   python3 scripts/render_design_preview.py design-specs/<deck-name>.md
   # Opens in browser: theme palette chips + per-slide wireframe thumbnails.
   # Variety warnings (3-streak layouts, no-diagram decks) appear at top.
   

3. Generate background images: Run the gradient background generator script

4. Create slides: Use PptxGenJS (Node.js) with the AWS theme constants

5. Add SVG visuals: Generate SVG diagrams for architecture/flow slides and embed as images

6. Apply animations: Design contextual animations and apply via apply_animations.py

7. QA (two-phase): QA uses two complementary layers — programmatic validation catches what rendered images hide (out-of-bounds shapes, font violations), then visual inspection catches what code cannot judge (aesthetics, readability).

   Phase 1 — Programmatic QA (fast, deterministic, run in main agent):

   python3 scripts/qa_validate.py output.pptx
   

   If critical issues are found (exit code 1), fix them before proceeding to Phase 2.

   Phase 2 — Visual QA (kiro preferred, subagent fallback): Delegates image-heavy inspection to a separate context to protect the main context window. Prefer kiro CLI (Opus 4.6) for higher quality analysis with severity classification. If kiro is not available, fall back to a subagent (Sonnet 4.6+).

# Step 1: Generate AWS gradient backgrounds
python3 scripts/create_aws_slide.py backgrounds --output-dir {workspace}/artifacts/myslide-assets/

# Step 2: Run the PptxGenJS creation script (generated per presentation)
node create_presentation.js

# Step 3: Apply animations (design JSON spec per presentation context)
python3 scripts/apply_animations.py output.pptx animations.json -o animated.pptx

# Step 4: Programmatic QA — catches structural issues renderers hide
python3 scripts/qa_validate.py output.pptx

# Step 5: Visual QA — prefer kiro (Opus 4.6), fall back to subagent (see below)

B. Editing an Existing Slide

When the user says "change slide 3" or "update the title slide":

1. Identify which slide(s) to modify
2. Read the current slide content (markitdown + image inspection)
3. Apply targeted changes (text, colors, layout, or visual elements)
4. Re-render and verify only the affected slides

B.1 Overlay on a File You Didn't Generate

If the customer (or a teammate) has drawn a slide in PowerPoint and you need to ADD a few elements to it without redrawing everything, use python-pptx overlay rather than regenerating from scratch. See references/pptx-overlay.md for the full workflow, including the add_connector endpoint-vs-width trap that is the most common cause of diagonal arrows piercing the slide title. That reference also has ready-to-copy helper functions for dashed lines, arrowheads, transparent container boxes, and zero-margin text labels.

Typical overlay use cases:

• Adding a new external system group (e.g., MCP bridge to on-premises systems) to an existing architecture slide
• Inserting a callout or annotation on a partner-provided deck
• Fixing one mispositioned label without touching anything else

For the MCP-specific styling (color, dash pattern, label convention), also read references/mcp-external-integration.md in the aws-diagram skill. Even when overlaying on an existing file, the visual conventions should match the from-scratch aws-diagram output so diagrams across the deck look consistent.

C. Sub-Agent Strategy for Large Decks (8+ slides)

For presentations with many slides, use parallel sub-agents to maximize throughput. Each sub-agent handles an independent group of slides.

Parallelization pattern:

• Agent 1: Title + Section Header slides (structural)
• Agent 2: Content slides (odd-numbered)
• Agent 3: Content slides (even-numbered)
• Agent 4: SVG diagram generation for all visual slides

# Sub-agent prompt template:
Generate slides [N] through [M] for the AWS presentation.
- Use the AWS theme from references/aws-theme.md
- Background images are at: {workspace}/artifacts/myslide-assets/
- Save individual slide JS snippets to: {workspace}/artifacts/myslide-parts/slide-{N}.js
- Follow the layout pattern specified for each slide type.

After all sub-agents complete, combine the JS snippets into one PptxGenJS script and execute.

Visual Diversity Strategy (CRITICAL)

Never repeat the same visual pattern more than twice in a deck. Slides that are all "dark box + bullet text" create visual fatigue. Use a mix of these patterns:

Hybrid Approach: SVG Visuals + Native PPTX Text

The most effective method combines SVG diagrams (for gradient shapes, glow effects, icons) with native PPTX text (for editability). The workflow:

1. Generate SVG with visuals only (shapes, gradients, glow, icons — NO text)
2. Convert to transparent PNG (remove background <rect> fill, set to fill="none")
3. Embed PNG as slide background image at the content area position
4. Overlay native PPTX text using slide.addText() positioned to match SVG element locations

// Step 1: SVG visual as background (no text, transparent bg)
slide.addImage({ data: noTextPngBase64, x: 0.15, y: 1.2, w: 13.0, h: 5.4 });

// Step 2: Native PPTX text on top (editable in PowerPoint)
slide.addText('Title', { x: 5.0, y: 3.0, w: 3.5, h: 0.5, fontSize: 24, bold: true, ... });

Benefits:

• User can edit text directly in PowerPoint
• Gradient/glow effects preserved from SVG
• Transparent PNG works on any slide background color
• Text is searchable and accessible

SVG Infographic Patterns (use svg-diagram skill)

Generate these via svg-diagram skill with transparent backgrounds:

Pattern	When to Use	Example Slides
Hub-Spoke	Central concept + related items	"What is X?" with Pain→Hub→Solution
Radial 5-node	5 features/capabilities around a center	"Why X?" feature highlights
Horizontal Icon Strip	Sequential or parallel items	"5 Advantages" with icon+label
Cross Quadrant	2x2 categorization	"4 Pain Points" with icons per quadrant
Donut Chart	Market share, proportions	Statistics with percentage breakdown
Timeline	Evolution, roadmap	"4 Stages of AI Coding"
Architecture	System/data flow	"LLM Gateway Architecture"
Process Flow	Step-by-step	"3-Step Onboarding"

Transparent SVG Background Rule

All SVG infographics MUST have transparent backgrounds so they work on any template.

When generating SVGs, ensure the first <rect> (background) has fill="none":

<rect width="1200" height="500" fill="none"/>

When stripping text from existing SVGs for hybrid approach:

svg = svg.replace(/<text[^>]*>[^<]*<\/text>/g, '');  // Remove all text elements

Gradient Shapes in PPTX

PptxGenJS doesn't support native gradient fills. Use pre-rendered gradient PNG images:

// Pre-generate gradient card backgrounds
async function createGradCard(w, h, c1, c2) {
  const svg = `<svg ...><linearGradient ...><rect fill="url(#g)"/></svg>`;
  return sharp(Buffer.from(svg)).png().toBuffer();
}
// Embed as image, then overlay transparent ROUNDED_RECTANGLE for border
slide.addImage({ data: gradPng, x, y, w, h });
slide.addShape(pres.shapes.ROUNDED_RECTANGLE, { x, y, w, h, fill: { type: 'none' }, line: { color, width } });

Build-Then-QA Workflow (MANDATORY)

Every batch of slide changes MUST be followed by visual QA before reporting completion.

1. Edit slide code (JS files)
2. Regenerate PPTX: node create_presentation.js
3. Convert to images: soffice → PDF → pdftoppm (target slides only)
4. Launch QA subagent (background): check alignment, readability, overlap
5. If QA finds issues → fix coordinates → regenerate → re-QA
6. Report completion only after QA PASS

For hybrid slides (SVG image + native text): QA must specifically check that PPTX native text aligns precisely with SVG visual elements (circles, boxes, etc). Coordinate misalignment is the #1 issue — text must be centered inside its target shape.

Fix workflow for misaligned text:

1. Open the user-modified PPTX (if available) with python-pptx
2. Extract actual coordinates: shape.left / 914400 (EMU to inches)
3. Update JS code with corrected coordinates
4. Regenerate and re-verify

Hybrid Slide Alignment Lessons (from user corrections)

These rules come from analyzing user manual corrections on hybrid slides. Apply these BEFORE generating to minimize manual fixes needed:

1. SVG Image Vertical Offset SVG infographic images should start below the title with extra margin:

• Title ends at y=1.0. Image should start at **y=1.51.6**, not y=1.2
• This gives breathing room and prevents cramped feeling

2. Radial/Spoke Layout Text Placement For radial (hub + spoke) diagrams:

• Top spoke: text goes ABOVE the icon circle (y = icon_top - 0.6)
• Left spokes: text goes LEFT of icon (x = icon_left - text_width - 0.1)
• Right spokes: text goes RIGHT of icon (x = icon_right + 0.1)
• Bottom spokes: text goes BELOW icon (y = icon_bottom + 0.1)
• Allow negative x values (e.g. x=-0.15) for left-edge spokes — PPTX clips gracefully

3. Horizontal Icon Strip Text Y-Position For 5-column horizontal layouts (icon above, text below):

• Icon occupies y=1.5~3.5 area
• Title text starts at y=3.83 (not 4.0) — tighter to icon bottom
• Korean subtitle at y=4.23, description at y=4.58
• Column spacing: calculate exact x from SVG icon centers, not evenly dividing

4. Quadrant Layout Text Positioning For cross-quadrant (2x2) layouts:

• Text x must be at least 0.15" right of icon right edge to prevent overlap
• User corrected: Q1/Q3 left columns x=2.67, Q2/Q4 right columns x=9.12~9.18
• Start y slightly below quadrant top: top row y=1.99, bottom row y=4.56~4.6

5. General SVG-Text Alignment Rule Calculate PPTX text positions FROM the SVG source coordinates:

pptx_x = (svg_element_cx / svg_viewBox_width) * slide_width
pptx_y = (svg_element_cy / svg_viewBox_height) * (image_h) + image_y_offset

Then fine-tune: add 0.1~0.2" margin away from icon edges.

QA Delegation Rules

Always delegate QA to subagents or kiro to protect the main context window:

• Visual QA (image-heavy) → run_in_background: true subagent or kiro skill
• Content fact-checking → dedicated subagent with reference MD files
• Alignment QA for hybrid slides → subagent with specific coordinate check instructions
• Never read slide images directly in the main agent context
• Main agent only receives QA text summary and applies fixes

Narrative Flow Patterns (Slide Ordering)

These patterns come from analyzing user reordering of customer case study slides. Plan slide order BEFORE generating based on the narrative pattern that fits.

Case Studies Pattern: "Specific → General" (Preferred)

When presenting customer case studies followed by a summary:

[Case 1] → [Case 2] → [Case 3] → [Section Header: "N 사례"] → [Summary Table]

Why this works:

• Audience sees concrete examples first (easier to grasp)
• Section header acts as a "conclusion marker" reinforcing the pattern
• Summary table at end provides reference/recap (not an introduction)

Avoid this order (common mistake):

[Section Header] → [Summary Table] → [Case 1] → [Case 2] → [Case 3]

This creates a "abstract first, concrete later" flow which feels academic and forces the audience to remember the table while watching individual cases.

Section Header Placement

Two valid placements:

1. Opening — "Here's what we'll cover" (introduces the section)
2. Closing/Bridging — "This is what we just saw" (wraps up the section, bridges to next)

The closing/bridging placement is especially effective when:

• Cases have been shown individually first
• The section serves as a summary or pivot point
• You want the audience to mentally organize what they just saw

Strength-of-Recommendation Language

User corrections consistently favor conditional recommendations over absolute ones, especially in customer-facing decks:

Too Strong (Avoid)	Preferred (Conditional)
"Amazon Bedrock 우선 검토 권장"	"엔터프라이즈 거버넌스 상이라면 → Amazon Bedrock 우선 검토 권장"
"X is the best choice"	"If your requirement is Y, X is the best choice"
"Always use Z"	"For scenarios requiring Z, use Z"

Frame recommendations with the condition/qualifier FIRST so the customer feels empowered rather than dictated to. This is especially important for AWS customer-facing content where neutrality matters.

Slide Types

Each presentation should use a MIX of these layouts. Never repeat the same layout more than twice in a row.

Type	When to Use	Reference
Title	First slide (single speaker)	slide-patterns.md > Title Slide
Title (Two Speakers)	First slide (co-presentation)	slide-patterns.md > Two-Speaker Title
Agenda	Second slide, table of contents	slide-patterns.md > Agenda Slide
Section Header	Chapter dividers (01, 02...)	slide-patterns.md > Section Header
Content Card	Key points with icon	slide-patterns.md > Content Card
Two Column Card	Side-by-side options/concepts	slide-patterns.md > Two Column Card
Three Column	Comparisons, 3 options	slide-patterns.md > Three Column
Process Flow	User scenarios, step-by-step flows	slide-patterns.md > Process Flow
Comparison Table	Option comparisons, feature matrices	slide-patterns.md > Comparison Table
Architecture	System diagrams	slide-patterns.md > Architecture
Venn/Comparison	Overlapping concepts	slide-patterns.md > Venn Diagram
Screenshot+Text	Demo/console walkthrough	slide-patterns.md > Screenshot
Summary Grid	2x2 key takeaways	slide-patterns.md > Summary Grid
Evolution/Progression	Maturity stages, AI evolution	slide-patterns.md > Evolution
Multi-Card Grid	2x2 or 3x2 feature cards	slide-patterns.md > Multi-Card Grid
Gradient Border Cards	Light cards with colored borders on dark bg	slide-patterns.md > Gradient Border Cards
Thank You	Last slide	slide-patterns.md > Thank You

Light Theme Additional Patterns

These layouts are specific to the Light theme (L100/field enablement style). See references/light-theme.md for full code examples.

Type	When to Use	Reference
Section Divider (Light)	Chapter breaks with gradient blob accents	light-theme.md > Section Divider
Data + Citation	Market data with bar charts + source quotes	light-theme.md > Data + Citation
Key Points Grid	2x3/2x4 grid with purple ALL-CAPS labels	light-theme.md > Key Points Grid
Feature Badges + Screenshot	Tan pill badges + product screenshot	light-theme.md > Feature List + Screenshot
Full-Color Background	Bold single-color slide (max 1 per deck)	light-theme.md > Full-Color Background Slide
Step Guide	Purple numbered badges + screenshots	light-theme.md > Step Guide with Screenshots
Customer Case Study	Logo + Challenge/Solution/Result + pill tags	light-theme.md > Customer Case Study
Do's vs Don'ts	Green/red color-coded comparison columns	light-theme.md > Do's vs Don'ts Comparison
CTA Slide	Orange numbered badges + action items	light-theme.md > Call to Action
Dashboard Evidence	Product UI screenshot as proof point	light-theme.md > Dashboard Screenshot Slide

SVG Visual Elements

For slides that need diagrams, architecture visuals, or flowcharts, generate inline SVG and convert to PNG for embedding. This dramatically improves visual quality over plain text slides.

When to generate SVGs:

• Architecture or system flow explanations
• Process/workflow diagrams
• Comparison charts or feature matrices
• Any slide where the content is inherently visual

SVG generation approach:

# Generate SVG with official AWS service icons, convert to PNG for embedding
python3 scripts/create_aws_slide.py svg-diagram \
  --type architecture \
  --elements "CloudFront,API Gateway,Lambda,Bedrock,S3" \
  --output {workspace}/artifacts/myslide-assets/arch-diagram.png

Available icon names (use these as element names for automatic icon embedding): Lambda, EC2, S3, DynamoDB, API Gateway, CloudFront, Bedrock, Bedrock AgentCore, ECS, EKS, RDS, Aurora, VPC, ELB, Route 53, CloudWatch, IAM, Cognito, SNS, SQS, Step Functions, EventBridge, Kinesis, SageMaker, Redshift, Athena, Glue, KMS, WAF, Shield, Fargate, ECR, AppSync, OpenSearch, ElastiCache, Secrets Manager, ACM, and 200+ more. See icons/ directory for the full list of available service icons.

AgentCore component-specific icons: 56 variants covering 9 AgentCore components (logo, ai-agent, runtime, gateway, identity, code-interpreter, observability, browser-tool, memory) across 4 color accents (teal, blue, purple, cyan) and 2 themes (light, dark). Pattern: agentcore-{component}-{color}-{theme}.svg — e.g., agentcore-runtime-teal-light.svg. Use these in slides when highlighting individual AgentCore services with matching color to slide theme:

• Light-theme (L100/field-enablement) slides → *-light variants (black outline + accent)
• Dark-theme (reInvent) slides → *-dark variants (white outline + accent) Embed via <img src="icons/agentcore-{name}.svg"> or place through svg-diagram.

Alternatively, craft SVG inline in the Node.js script and use sharp to convert to PNG base64.

Diagram Type Selection for Presentations

Presentations almost always need High-Level diagrams (logical grouping), not Infrastructure diagrams (VPC/Subnet).

Presentation Context	Diagram Type	Key Characteristics
Executive briefing, sales pitch	High-Level	Hub-spoke layout, generic containers, no VPC/Subnet
Technical deep-dive, Well-Architected review	Infrastructure	VPC/AZ/Subnet nesting, security boundaries
Platform overview (e.g., AgentCore, EKS)	High-Level (Hub-Spoke)	Central runtime + radiating spoke services
Event-driven / microservices overview	High-Level	EventBridge as central bus, producers left, consumers right

High-Level Architecture Pattern (Hub-Spoke)

For presentation slides, the hub-spoke pattern is most effective:

• Central service (Bedrock, EventBridge, EKS) at the center of the diagram
• Spoke services radiate outward with clear directional arrows
• Logical grouping using generic container (dashed border) instead of VPC/Subnet
• Observability (CloudWatch, X-Ray) at the bottom, separated from main flow
• No AWS Cloud outer container -- cleaner for slides
• Multi-directional arrows -- not just left-to-right; use top/bottom/left/right freely
• Descriptive arrow labels -- "Streamable HTTP", "IdP integration", "Metrics & logs"

Arrow-Icon Collision Rules

• Arrows must NEVER pass through another service box. This is the #1 architecture diagram defect. Before drawing each arrow, trace the path from source to target and verify no intermediate box lies in the way. If one does:
  1. Route around: Use two LINE segments (L-shape or Z-shape) to go around the box
  2. Move the blocking box to a row/column that clears the path
  3. Connect via the intermediate box if there's a logical flow through it
• Arrow segments must maintain >= 10px clearance from any icon bounding box
• In hub-spoke layouts, radial arrows must not cross sibling spokes -- stagger spoke y-positions
• Validation: After all elements are placed, mentally trace every arrow and confirm it does not cross any box. See slide-patterns.md > Arrow Routing Rules for code examples.

aws-diagram Skill Integration

For complex architecture diagrams (VPC nesting, orthogonal arrows, 8+ services), use the aws-diagram skill instead of create_aws_slide.py. It produces native PPTX architecture slides that can be merged into myslide decks:

# 1. Generate architecture diagram with aws-diagram skill
python3 /path/to/aws-diagram/scripts/generate_diagram.py \
  -i diagram.json -o {workspace}/artifacts/myslide-assets/arch.svg --png --pptx {workspace}/artifacts/myslide-assets/arch-slide.pptx

# 2. Merge into myslide deck using add_slide.py
python3 scripts/add_slide.py --deck output.pptx --insert {workspace}/artifacts/myslide-assets/arch-slide.pptx --position 4

For High-Level diagrams in aws-diagram JSON, use "type": "generic" containers:

{"id": "platform", "type": "generic", "label": "Amazon Bedrock AgentCore",
 "children": ["runtime", "code-interp", "identity", "memory"]}

SVG layer order rule (applies to all SVG diagram generation): Icons must render ABOVE arrows. Render order: background > containers > arrows > callouts > icons.

Image Generation (Optional)

When a slide needs a conceptual illustration, hero image, or visual metaphor that cannot be expressed with SVG diagrams or AWS icons, use the sd35l skill (GA) to generate images via Amazon Bedrock. nova2-omni is also available as an alternative (gated preview).

Requires: If used, the sd35l skill must exist under {WORKING_DIR}/skills/sd35l (or adjust the find path above).

When to Use Image Generation vs SVG Diagrams

Content Type	Use SVG/Icons	Use sd35l
AWS architecture diagrams	Yes	No
Process flows, step diagrams	Yes	No
Conceptual hero images (AI brain, cloud, etc.)	No	Yes
Abstract background visuals	No	Yes
Product/scenario illustrations	No	Yes
Screenshot placeholders	No	Yes
Data flow with service icons	Yes	No

Slide-Optimized Aspect Ratios

Use Case	Aspect Ratio	Slide Coverage
Full-slide background	16:9	Entire slide behind text
Hero image (title slide)	16:9	Right 60-70% of slide
Half-slide illustration	2:3 or 3:4	Left/right half
Card illustration	1:1	Inside a content card
Banner (wide strip)	21:9	Top or bottom strip

Integration Workflow

# 1. Generate image with sd35l (resolve under application/skills only)
SD35L_SCRIPT=$(find "${WORKING_DIR}/skills" -path "*/sd35l/scripts/generate_image.py" 2>/dev/null | head -1)

python3 "$SD35L_SCRIPT" \
  --prompt "Abstract dark gradient with glowing neural network connections, deep navy and purple tones, futuristic technology atmosphere, minimal clean composition" \
  --negative-prompt "text, watermarks, logos, people, bright colors, white background" \
  --aspect-ratio 16:9 \
  --seed 42 \
  --output-dir {workspace}/artifacts/myslide-assets/

# 2. Result JSON: {"model": "...", "seed": 42, "images": ["{workspace}/artifacts/myslide-assets/sd35l_1.png"]}

# 3. Embed in PptxGenJS using base64

Prompt Guidelines for Presentation Images

Style keywords to include:

• "dark background", "deep navy", "dark gradient" (matches AWS theme)
• "minimal", "clean composition" (professional look)
• "glowing", "luminous accents" (matches orange/magenta highlights)
• "futuristic", "technology", "digital" (AWS tech context)

Standard negative prompt for all slide images:

"text, watermarks, logos, bright white background, oversaturated, cartoon, cluttered, busy composition, blurry"

Prompt templates by slide type:

Slide Type	Prompt Pattern
Title hero	"[Topic concept] visualization, dark futuristic background, glowing [accent color] accents, cinematic wide shot, professional technology illustration"
Content illustration	"[Concept] depicted as [visual metaphor], dark navy background, clean minimal style, soft ambient lighting, 3D render"
Background overlay	"Abstract [theme] pattern, dark gradient from deep navy to black, subtle glowing particles, seamless texture, minimal"
Card thumbnail	"[Subject icon/symbol], centered on dark background, simple flat design with glow effect, single color accent, square composition"

Embedding Generated Images in PptxGenJS

const fs = require('fs');

const heroImage = fs.readFileSync('{workspace}/artifacts/myslide-assets/sd35l_1.png');
const heroBase64 = 'image/png;base64,' + heroImage.toString('base64');

// Full-slide background image
slide.background = { data: heroBase64 };

// Or position as a visual element
slide.addImage({
  data: heroBase64,
  x: 5.5, y: 0, w: 7.83, h: 7.5,  // Right 60% of slide
});

// Semi-transparent overlay on top of image (for text readability)
slide.addShape(pres.shapes.RECTANGLE, {
  x: 0, y: 0, w: 13.33, h: 7.5,
  fill: { color: "000000", transparency: 50 }
});

Cost Awareness

• SD3.5 Large: ~$0.04/image regardless of aspect ratio
• A typical 10-slide deck with 3-4 generated images: ~$0.16
• Use --seed to reproduce exact images when iterating

See references/image-generation-integration.md for detailed prompt recipes and advanced techniques.

Rounded Rectangle Default

All card-like shapes (content cards, agenda items, grid cells, tag badges, summary bars) MUST use ROUNDED_RECTANGLE with rectRadius instead of plain RECTANGLE. This gives a modern, polished look consistent with contemporary UI design.

Radius Guidelines

Element	rectRadius	Notes
Large cards (content, 2-col, 3-col)	0.10 - 0.12	Main content containers
Inner boxes (code blocks, option cards)	0.06 - 0.08	Nested within larger cards
Tag badges (effort/impact labels)	0.15	Pill-shaped badges
Progress bars	0.10	Pill-shaped bars
Summary/footer bars	0.08	Horizontal wide bars

Accent Styling with Rounded Corners

Do NOT overlay thin rectangular accent bars on ROUNDED_RECTANGLE shapes (they won't align with corners). Instead, use line border property for accent colors:

// ✅ CORRECT: Rounded card with colored border accent
slide.addShape(pres.shapes.ROUNDED_RECTANGLE, {
  x: 0.5, y: 1.2, w: 5.8, h: 4.5, rectRadius: 0.12,
  fill: { color: "161E2D" },
  line: { color: "C91F8A", width: 1.5 },  // accent color as border
  shadow: mkShadow()
});

// ❌ WRONG: Overlay accent bar on rounded corners
slide.addShape(pres.shapes.ROUNDED_RECTANGLE, { ... });
slide.addShape(pres.shapes.RECTANGLE, { x: 0.5, y: 1.2, w: 0.08, h: 4.5, fill: { color: "C91F8A" } });

When to Keep RECTANGLE

Only use plain RECTANGLE for:

• Full-slide overlays (dark transparency over hero images)
• Table cell backgrounds (must tile without gaps)
• Decorative gradient accent bars at slide edges

Thin Accent Lines Under Titles (DO NOT USE)

Do NOT add thin orange/pink accent lines (h: 0.04) under slide titles. This creates visual noise and looks repetitive across slides. The title text itself is sufficient as the visual anchor. If visual separation is needed between title and content, use whitespace (0.3"+ gap) instead of a line.

// ❌ WRONG: Thin accent line under title
slide.addShape("rect", {
  x: 0.8, y: 1.05, w: 1.5, h: 0.04, fill: { color: C.orange },
});

// ✅ CORRECT: Use whitespace gap between title and content
// Title at y: 0.33, content starts at y: 1.41 (natural 0.3"+ gap)

Color Discipline (CRITICAL)

Slides look more professional with fewer, well-chosen colors. Too many colors create visual noise and distract from the message. Limit each slide to 5 core colors maximum. A restrained palette feels intentional and polished; a rainbow of colors feels chaotic.

The 5 Core Colors (Dark Theme)

Role	Color	HEX	When to use
Background	Deep Purple-Black	09051B	Slide background (via gradient image)
Primary Text	White	FFFFFF	All headings, body text, bullets
Emphasis	Orange	F66C02	Key terms, highlights, numbered badges
Container	Dark Navy	161E2D	Card fills, table cells, box backgrounds
Secondary	Light Slate	C8D0D8	Subtitles, descriptions, captions (readable on projectors)

The 5 Core Colors (Light Theme)

Role	Color	HEX	When to use
Background	White	FFFFFF	Slide background
Primary Text	Near Black	1A1A1A	All headings, titles
Accent	Sky Blue	4FC3F7	Table headers, bullets, links
Card Fill	Cream/Beige	F5F0EB	Card backgrounds, content areas
Body Text	Dark Gray	333333	Body text, descriptions

Light theme allows additional accents sparingly: coral (C96842) for key stats, purple (6B46C1) for architecture labels/badges, AWS orange (FF9900) for internal badges and CTA numbers. See light-theme.md for full palette.

Allowed Accent (sparingly)

• C91F8A (Magenta) — card border lines only, max 1-2 per slide
• 5600C2 (Purple) — already in the gradient background, no need to add separately

Colors to AVOID in regular slides

These colors exist in the theme file for special cases but should NOT appear in normal content slides. Using them creates a cluttered, inconsistent look:

• FF28EF (Neon Pink) — only for section header numbers
• ABABE3 (Lavender), FF9EA2 (Salmon Pink) — only for complex architecture diagrams
• 00A0C8 (Teal), 69AE35 (Green) — only when semantically meaningful (e.g., success/info)
• 010135, 02043B — only in multi-layer architecture diagrams

Rule of thumb: If you're about to use a 6th color, stop and ask whether one of the 5 core colors can serve the same purpose.

Gradient Fills for Cards and Containers

PptxGenJS does not support native gradient fills on shapes. To create gradient card backgrounds (e.g., a subtle dark-to-darker gradient inside a card), render an SVG gradient rectangle and convert to PNG:

const sharp = require('sharp');

// Create a gradient card background (e.g., darkNavy to bgBase)
async function createGradientCard(w, h, colors = ['161E2D', '09051B'], direction = 'vertical') {
  const pxW = Math.round(w * 96);  // inches to pixels at 96dpi
  const pxH = Math.round(h * 96);
  const [c1, c2] = colors;
  const gradDir = direction === 'vertical'
    ? 'x1="0" y1="0" x2="0" y2="1"'
    : 'x1="0" y1="0" x2="1" y2="0"';

  const svg = `<svg xmlns="http://www.w3.org/2000/svg" width="${pxW}" height="${pxH}">
    <defs>
      <linearGradient id="g" ${gradDir}>
        <stop offset="0%" stop-color="#${c1}"/>
        <stop offset="100%" stop-color="#${c2}"/>
      </linearGradient>
    </defs>
    <rect width="${pxW}" height="${pxH}" rx="12" fill="url(#g)"/>
  </svg>`;

  const png = await sharp(Buffer.from(svg)).png().toBuffer();
  return 'image/png;base64,' + png.toString('base64');
}

// Usage: gradient card as image background
const gradBg = await createGradientCard(12.5, 4.84, ['161E2D', '0D1117']);
slide.addImage({ data: gradBg, x: 0.42, y: 1.41, w: 12.5, h: 4.84 });

// Then add text/shapes on top of the gradient card
slide.addText("Title", { x: 0.8, y: 1.8, w: 11.5, h: 0.5, ... });

Gradient presets for common use cases:

Use Case	From	To	Direction
Card background	161E2D	0D1117	vertical (top→bottom)
Header bar	161E2D	09051B	horizontal (left→right)
Highlight card	1A0B3D	161E2D	vertical
Summary footer	0D1117	161E2D	horizontal

Use gradient fills sparingly — they add visual depth but overuse diminishes the effect. One or two gradient cards per presentation is ideal.

Gradient Borders (for any shape)

Gradient borders work on any card — single Content Card, Multi-Card Grid, or the Gradient Border Cards layout. The technique: render an SVG with a gradient-filled rectangle and a smaller inner rectangle in the fill color, creating a border effect.

// Reusable gradient border generator — works for any card size
async function createGradientBorder(w, h, borderColors, fillColor = 'F2F4F4', borderWidth = 3) {
  const pxW = Math.round(w * 96);
  const pxH = Math.round(h * 96);
  const [c1, c2] = borderColors;
  const r = 12;  // corner radius

  const svg = `<svg xmlns="http://www.w3.org/2000/svg" width="${pxW}" height="${pxH}">
    <defs>
      <linearGradient id="b" x1="0" y1="0" x2="0" y2="1">
        <stop offset="0%" stop-color="#${c1}"/>
        <stop offset="100%" stop-color="#${c2}"/>
      </linearGradient>
    </defs>
    <rect width="${pxW}" height="${pxH}" rx="${r}" fill="url(#b)"/>
    <rect x="${borderWidth}" y="${borderWidth}"
          width="${pxW - borderWidth * 2}" height="${pxH - borderWidth * 2}"
          rx="${r - 1}" fill="#${fillColor}"/>
  </svg>`;

  const png = await sharp(Buffer.from(svg)).png().toBuffer();
  return 'image/png;base64,' + png.toString('base64');
}

// Example: Single Content Card with orange→magenta gradient border
const cardFrame = await createGradientBorder(12.5, 4.84, ['F66C02', 'C91F8A'], '161E2D');
slide.addImage({ data: cardFrame, x: 0.42, y: 1.41, w: 12.5, h: 4.84 });
// Then add text on top...

// Example: Light card with purple→blue gradient border
const lightCard = await createGradientBorder(5.8, 4.5, ['5600C2', '2D7CFB'], 'F2F4F4');
slide.addImage({ data: lightCard, x: 1.0, y: 1.5, w: 5.8, h: 4.5 });

Common gradient border presets:

Style	From	To	Fill	Effect
Warm accent	F66C02	C91F8A	161E2D	Orange→Magenta on dark card
Cool accent	5600C2	2D7CFB	F2F4F4	Purple→Blue on light card
Brand gradient	C91F8A	5600C2	161E2D	Magenta→Purple on dark card
Subtle	C8D0D8	161E2D	161E2D	Gray fade, subtle

See slide-patterns.md > Gradient Border Cards for the full multi-card layout pattern.

Typography Size Rules (CRITICAL)

Body text must NEVER be smaller than 15pt. This is the #1 visual quality issue. Table cells, comparison columns, and process flow card text are NOT exceptions — they must also be 15pt minimum. Only footer copyright and slide-number captions may go below 15pt.

Minimum Font Size by Element

Element	Min Size (pt)	Recommended (pt)	Weight
Slide title	36	36-44	Bold/Heavy
Section number (01, 02)	36	36	Bold
Card title / Sub-header	20	20-24	Bold
Body text	15	16	Regular
Bullet items	15	15-16	Regular
Table cell text	15	15-16	Regular
Process flow card text	15	15-16	Regular
Three-column body	15	15-16	Regular
Caption / Footer / Copyright	8	8-10	Light

The only elements allowed below 15pt are:

• Footer copyright text (8pt)
• Slide number labels (8-10pt)
• Service name labels under icons in architecture diagrams (10-12pt)

Everything else — including table cells, comparison columns, process flow descriptions, multi-card grid body text — MUST be 15pt or higher. If content overflows at 15pt, reduce the text content rather than shrinking the font.

Card-to-Text Balance Rule

Text should fill at least 60% of the card's vertical space. If a card has excessive bottom whitespace (>30% empty), apply one or more of these fixes:

1. Increase font size (first choice) — bump body text to 16-18pt
2. Increase line spacing — use lineSpacingMultiple: 1.3 to 1.5
3. Reduce card height — shrink the card to fit the content snugly
4. Add vertical centering — use valign: "middle" on text boxes

// WRONG: Small text in large card = excessive whitespace
slide.addText(bodyText, {
  x: 1, y: 2.5, w: 5, h: 3.5,
  fontSize: 12, // TOO SMALL
});

// CORRECT: Properly sized text filling the card
slide.addText(bodyText, {
  x: 1, y: 2.5, w: 5, h: 3.5,
  fontSize: 16,
  lineSpacingMultiple: 1.3,
  valign: "top",
});

Grid/Multi-Card Slides

For 2x2 grids and 3-column layouts where space is tighter:

• Card title: 18-20pt bold
• Body text: 15pt minimum (never drop to 12-14pt)
• Table cells: 15pt minimum (not 13-14pt — reduce column count or text instead)
• Process flow card text: 15pt minimum
• If content overflows, reduce text content rather than font size
• Use paraSpaceAfter: 6 instead of paraSpaceAfter: 10 to reclaim space

Text Emphasis Rules

The AWS template uses specific color highlighting for emphasis:

• Orange (F66C02): Primary emphasis for key terms, important concepts
• Magenta (C91F8A): Secondary emphasis, links, interactive elements
• White bold: Section headers within content areas
• Regular white: Body text

Use rich text arrays in PptxGenJS to apply selective highlighting:

slide.addText([
  { text: "Customer data is ", options: { color: "FFFFFF", fontSize: 14 } },
  { text: "never used for training", options: { color: "F66C02", fontSize: 14, bold: true } },
  { text: " any model.", options: { color: "FFFFFF", fontSize: 14 } }
], { x: 1, y: 2, w: 8, h: 1 });

Gradient Effects

PptxGenJS doesn't natively support gradients on text or shape borders. Two workarounds are documented in references/pptxgenjs.md:

• Gradient Text (section "Gradient Text"): Render as SVG, convert to PNG via sharp, embed as image. Good for large hero numbers like "80%".
• Gradient Border (section "Gradient Border"): Post-process the generated PPTX with python-pptx to inject OOXML gradient line XML. This is the preferred approach for shape borders because the result is native PowerPoint and remains editable. Use apply_gradient_border() from the reference.

Conversational Editing

The user may request changes like:

• "3번 슬라이드 제목을 바꿔줘" → Read slide 3, modify title text
• "아키텍처 다이어그램을 추가해줘" → Generate SVG + add new slide
• "색상을 좀 더 밝게" → Adjust fill/text colors
• "순서를 바꿔줘" → Reorder slides in presentation.xml

For each edit request:

1. Show the user what the current slide looks like (render to image)
2. Describe the proposed change
3. Apply the change
4. Show the result for confirmation

Animations (Contextual Design)

Animations are designed per-presentation based on content and narrative flow. There are no fixed templates — each slide gets animations tailored to its story.

Read references/animations.md for the full primitives reference.

Workflow

1. Design animations contextually after slides are created:

   • What appears first? (title, framing context)
   • What's the focal sequence? (data flow, process steps)
   • What's the conclusion? (output, call-to-action)

2. Write JSON spec describing the animation sequence per slide

3. Apply with script:

   # List shapes to find target names/IDs
   python3 scripts/apply_animations.py deck.pptx --list-shapes
   
   # Apply animations
   python3 scripts/apply_animations.py deck.pptx animations.json -o animated.pptx
   

Key Primitives

Category	Effects	Use Case
Entrance	appear, fade_in, fly_in, wipe, zoom_in	Revealing content step-by-step
Emphasis	pulse, spin, grow_shrink, color_pulse	Drawing attention to key element
Exit	disappear, fade_out, fly_out, zoom_out	Removing elements
Motion	motion_path (SVG path syntax)	Data flow, pipeline visualization
Property	property_change (fill_color, visibility)	State transitions

Triggers

Trigger	Behavior
onClick	New click point — presenter controls timing
withPrevious	Simultaneous with previous animation
afterPrevious	Auto-starts after previous completes

Shape Naming

PptxGenJS does NOT preserve custom name attributes. After generating the PPTX, run --list-shapes to get actual shape names/IDs, then build the animation JSON spec using those names.

Animation Design Principles

• Less is more — animate only what serves the narrative
• Consistent timing — use 500ms for standard, 300ms for fast, 800-1000ms for dramatic
• Group related elements — card + text should animate together (withPrevious)
• Build left-to-right — for pipelines and data flows
• Fade for text, Fly for shapes — general heuristic for visual harmony

Team-Up Strategy (Large Presentations, 15+ slides)

For very large or complex presentations, use the team-assemble pattern to coordinate multiple agents working in parallel. This is more powerful than simple sub-agents because it enables inter-agent communication, shared task tracking, and phased execution.

When to use Team-Up instead of Sub-Agents:

• 15+ slides with diverse content types
• Presentations requiring research + design + implementation phases
• When SVG diagram generation is complex and needs dedicated attention

Team Composition:

Agent	Role	Model	Responsibility
lead	Orchestrator	opus	Plan structure, assign tasks, final QA
designer	Visual Design	sonnet	Generate backgrounds, SVG diagrams, icons
writer-1	Content (slides 1-N/2)	sonnet	Write first half slide code
writer-2	Content (slides N/2+1-N)	sonnet	Write second half slide code
qa	Quality Assurance	opus/sonnet	Run qa_validate.py first, then visual QA via kiro (Opus) or subagent (Sonnet 4.6+)

Team Workflow:

1. lead: Analyze topic, create slide outline, assign tasks
2. designer + writer-1 + writer-2: Work in parallel
   - designer generates all background/SVG assets
   - writers create PptxGenJS code for assigned slides
3. lead: Combine all parts into single presentation script
4. qa: Run qa_validate.py, fix critical issues, then render and inspect visually
5. lead: Apply fixes, finalize

Task Assignment Pattern:

TaskCreate: "Generate AWS gradient backgrounds and SVG diagrams"
  -> owner: designer
TaskCreate: "Write slides 1-8 (Title, Section, Content)"
  -> owner: writer-1, blockedBy: [designer task]
TaskCreate: "Write slides 9-15 (Architecture, Summary, Thank You)"
  -> owner: writer-2, blockedBy: [designer task]
TaskCreate: "QA all rendered slides"
  -> owner: qa, blockedBy: [writer-1 task, writer-2 task]

Bundled Scripts Reference

All scripts are self-contained within scripts/:

Script / Directory	Purpose
icons/	304 official AWS service icons (SVG), including 56 AgentCore component variants
scripts/create_aws_slide.py	Generate AWS gradient backgrounds, SVG diagrams with official icons, logo
scripts/office/soffice.py	Convert PPTX to PDF via LibreOffice
scripts/office/unpack.py	Unpack PPTX to XML for direct editing
scripts/office/pack.py	Repack edited XML back to PPTX
scripts/thumbnail.py	Generate thumbnail grid for visual overview
scripts/clean.py	Clean PPTX XML (remove unused elements)
scripts/add_slide.py	Add slides to existing PPTX
scripts/apply_animations.py	Inject OOXML animations from JSON spec into PPTX
scripts/qa_validate.py	Programmatic QA — bounds, connectors, font sizes, zero-size shapes
scripts/render_design_preview.py	Design spec → HTML preview (palette chips + slide wireframes) for pre-build approval
references/design-spec-template.md	Pre-build design spec template (slide table, approval flow, variety check)
references/image-generation-integration.md	sd35l / nova2-omni image generation guide for slides

Two-Phase QA (ALWAYS run both phases):

Phase 1 — Programmatic QA (run in main agent, fast):

python3 "${WORKING_DIR}/skills/myslide/scripts/qa_validate.py" {pptx_path}
# Exit code 0 = pass, 1 = critical issues, 2 = error
# Add --json for machine-readable output
# Add --strict to include INFO-level findings

This catches what rendered images HIDE — shapes extending beyond slide boundaries, connector endpoints outside the slide, text below 15pt, zero-size shapes. These issues are invisible in rendered images because renderers clip at the slide edge. Fix all critical issues before proceeding to Phase 2.

Phase 2 — Visual QA (kiro preferred, subagent fallback):

Visual QA is image-heavy and consumes significant context window tokens. Delegate it to a separate context — never read slide images in the main agent.

Option A — Kiro CLI (preferred, Opus 4.6):

Kiro uses Opus 4.6 which produces more structured reports with severity classification and deeper design system analysis. Use when kiro CLI is installed.

# First convert PPTX to images
python3 "${WORKING_DIR}/skills/myslide/scripts/office/soffice.py" --headless --convert-to pdf {pptx_path}
pdftoppm -jpeg -r 150 {pdf_path} {workspace}/artifacts/myslide-qa/slide

# Then delegate visual QA to kiro (script under application/skills/kiro if present)
KIRO_RUN=$(find "${WORKING_DIR}/skills" -path "*/kiro/scripts/run_kiro.sh" 2>/dev/null | head -1)
# If KIRO_RUN is empty, use Option B below — do not run bash with an empty path.
bash "$KIRO_RUN" --trust-all --timeout 300 \
  "You are a visual QA inspector for AWS-themed PowerPoint presentations.

Read each slide image and inspect for visual quality:
$(for f in {workspace}/artifacts/myslide-qa/slide-*.jpg; do echo "- $f"; done)

Design system: dark bg (#09051B), white text, orange (#F66C02) emphasis,
dark navy (#161E2D) cards, rounded rectangles, body text min 15pt.

Phase 1 (programmatic) already verified: bounds, connectors, font sizes.
Focus on VISUAL quality only:
- Color contrast, orange highlight usage (max 2-3 per slide)
- Layout variety (no 3+ identical patterns), card whitespace (text >= 60%)
- SVG diagram clarity, icon layer order, arrow-icon clearance
- Thank You slide text overlap, generated image dark theme match

Per-slide report: PASS/FAIL with severity (경미/중간/심각) and specific issues.
Summary at end."

Option B — Subagent (fallback, Sonnet 4.6+):

Use when kiro is not available. Spawn a dedicated subagent for visual inspection.

You are a visual QA inspector for AWS-themed PowerPoint presentations.

Phase 1 (programmatic) already passed. Structural checks are verified.
Focus on VISUAL quality only.

1. Convert PPTX to images:
   python3 "${WORKING_DIR}/skills/myslide/scripts/office/soffice.py" --headless --convert-to pdf {pptx_path}
   pdftoppm -jpeg -r 150 {pdf_path} {output_prefix}

2. Read each slide image and check:
   - Color contrast sufficient (white on dark)
   - Orange highlights used sparingly (max 2-3 per slide)
   - Layout variety across deck (no 3+ identical patterns in a row)
   - SVG diagrams render cleanly without text overlap
   - Icons render above arrows (correct layer order)
   - No arrow lines crossing over service icons
   - Card whitespace ratio: text fills >= 60% of card height
   - Thank You slide: English/Korean text not overlapping
   - Generated images match dark theme (dark with accent glows)
   - Text overlay on images has sufficient contrast

3. Return a text-only report:
   - Per-slide: PASS or FAIL with specific issues
   - Summary: total pass/fail count and critical issues to fix

Do NOT return images. Only return text findings.

Choosing between Kiro and Subagent:

	Kiro (Opus 4.6)	Subagent (Sonnet 4.6)
Model quality	Higher (better design judgment)	Good (catches major issues)
Report structure	Severity levels, tables	Prose descriptions
Context isolation	Separate process	Subagent within Claude Code
Availability	Requires kiro CLI installed	Always available
Speed	Comparable	Comparable
Best for	Final QA before delivery	Quick iteration checks

Editing workflow (unpack/repack):

python3 scripts/office/unpack.py input.pptx unpacked/
# ... edit files in unpacked/ ...
python3 scripts/clean.py unpacked/
python3 scripts/office/pack.py unpacked/ output.pptx

QA Checklist

Phase 1 — Programmatic (qa_validate.py, run automatically)

These are checked by the script — no manual inspection needed:

• All shapes within slide boundaries (no out-of-bounds objects)
• All connector/arrow endpoints within slide boundaries
• No zero-length connectors (invisible arrows)
• No zero-size shapes (invisible objects)
• Body text fontSize >= 15pt (NEVER 12-14pt) — includes table cells, process flow cards, column text
• No text below 8pt (absolute minimum for captions/footers)

Phase 2 — Visual (subagent inspection)

These require human/AI visual judgment on rendered images:

• Every slide has a visual element (no text-only slides)
• Color contrast is sufficient (white text on dark backgrounds)
• Orange highlights are used sparingly (max 2-3 per slide)
• Layouts vary across the deck (no 3+ identical patterns in a row)
• Presenter info matches defaults or user-provided values
• All SVG diagrams render cleanly without text overlap
• SVG layer order correct: icons render above arrows (not obscured)
• No arrow lines cross over service icons (>= 10px clearance)
• Architecture diagrams use appropriate type (High-Level for presentations, Infrastructure for tech deep-dives)
• Hub-spoke diagrams have staggered spoke y-positions (no radial arrow crossings)
• Arrow markers use Open Arrow style (stroke, not filled polygon)
• All card/container shapes use ROUNDED_RECTANGLE (not plain RECTANGLE)
• Accent colors use line border, not overlay bars
• Max 5 colors per slide (bgBase, white, orange, darkNavy, lightSlate C8D0D8 + magenta border)
• No unnecessary colors (lavender, salmon, teal, green reserved for special cases only)
• Card whitespace ratio: text fills >= 60% of card height (no excessive bottom gaps)
• Grid/multi-card slides use 15pt+ body text (not smaller to "fit")
• Thank You slide: English/Korean text must not overlap (separate lines with sufficient spacing)
• Architecture diagram: every service box must include an AWS service icon (no empty boxes)
• Footer with AWS logo placement is consistent
• Nova Canvas images match AWS dark theme (predominantly dark with accent glows)
• Text overlay on generated images has sufficient contrast (dark overlay applied)
• Generated image dimensions match target slide area (no stretching/distortion)