agentsclimarketplace

Shipkit communications

Skill stefan-stepzero/shipkit/install/skills/shipkit-communications

Use when user wants to visualize or present shipkit content as HTML. Triggers: 'create presentation', 'visualize this', 'HTML report'.From its SKILL.md

Install
npx -y skills add stefan-stepzero/shipkit --skill shipkit-communications

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

14.3 KB, ~3.5k tokens by cl100k_base, as published. Nobody here has run it

shipkit-communications - Visual HTML Reports

Purpose: Create beautiful, visual HTML reports from any ShipKit Lite content for presentations, sharing, or visual review.

What it does: Reads markdown content from .shipkit/, generates styled HTML with purple gradient theme, archives old reports, creates latest.html.


When to Invoke

User triggers:

  • "Visualize the architecture decisions"
  • "Create HTML report"
  • "Make a visual status report"
  • "Generate presentation from project status"
  • "Show me quality checklist as HTML"
  • "Create shareable report"

Use cases:

  • Share architecture decisions with team
  • Present project status to stakeholders
  • Visual review of quality checklist
  • Export component documentation as HTML
  • Create visual specs for review

Prerequisites

Optional (depends on what user wants to visualize):

  • .shipkit/architecture.json - For architecture decisions
  • .shipkit/codebase-index.json - For component/route docs
  • .shipkit/stack.json - For tech stack overview
  • .shipkit/specs/{todo,active}/*.json - For specs
  • .shipkit/plans/{todo,active}/*.json - For implementation plans

No strict prerequisites - Can create HTML from any content.


Process

Step 1: Ask What to Visualize

Before generating anything, ask user what they want to visualize:

Present options:

What would you like to visualize as HTML?

1. Architecture Decisions - Visual timeline of decisions
2. Project Status - Complete health dashboard
3. Quality Checklist - Pre-ship verification
4. Component Documentation - All components/routes
5. Feature Specs - Active specifications
6. Implementation Plans - Current plans
7. Tech Stack - Visual stack overview
8. Custom - You specify the files

Choose a number or describe what you need:

Wait for user response.


Step 2: Determine Source Files

Based on user choice, identify which .shipkit/ files to read:

ChoiceFiles to ReadDescription Word
1. Architecture Decisions.shipkit/architecture.jsonarchitecture-decisions
2. Project StatusGlob .shipkit/**/*.jsonproject-health
3. Quality Checklist.shipkit/codebase-index.json, .shipkit/specs/{todo,active}/*.jsonquality-checklist
4. Component Documentation.shipkit/codebase-index.jsoncomponent-docs
5. Feature Specs.shipkit/specs/{todo,active}/*.jsonfeature-specs
6. Implementation Plans.shipkit/plans/{todo,active}/*.jsonimplementation-plans
7. Tech Stack.shipkit/stack.jsontech-stack
8. CustomUser-specified filesUser-specified description

Description word: Used for archive filename (2-4 kebab-case words)


Step 3: Read Source Content

Use Read tool to load source markdown files.

FOR COMPREHENSIVE REPORTS (Project Status), USE PARALLEL READS:

Launch these Read operations IN PARALLEL (single message, multiple tool calls):

1. Read: .shipkit/architecture.json
2. Read: .shipkit/stack.json
3. Read: .shipkit/codebase-index.json
4. Read: .shipkit/schema.json
5. Glob + Read: .shipkit/specs/{todo,active}/*.json
6. Glob + Read: .shipkit/plans/{todo,active}/*.json

Why parallel: All file reads are independent - no dependencies between them. Parallel reads significantly speed up comprehensive report generation.

Examples for single-source reports:

For architecture decisions:

Read: .shipkit/architecture.json

For quality checklist:

Read: .shipkit/codebase-index.json
Glob: .shipkit/specs/{todo,active}/*.json

IMPORTANT: Actually read files - don't generate placeholder content.


Step 4: Check for Existing latest.html

Check if .shipkit/communications/latest.html already exists:

If latest.html exists:
  1. Get current timestamp: YYYYMMDD-HHMM format
  2. Create archive filename: YYYYMMDD-HHMM-{description}.html
     Example: 20251228-1430-architecture-decisions.html
  3. Use Read tool to read latest.html content
  4. Use Write tool to create archive:
     Path: .shipkit/communications/archive/{timestamp}-{description}.html
     Content: [contents of old latest.html]
  5. Proceed to generate new latest.html

If latest.html does NOT exist:
  1. Proceed directly to generate new latest.html

Archive naming examples:

  • 20251228-1430-architecture-decisions.html
  • 20251228-1515-project-health.html
  • 20251228-1620-quality-checklist.html

Step 5: Generate HTML

Create beautiful, styled HTML using the template structure.

Complete HTML template: See references/html-template.md

Key template features:

  • Purple gradient theme (#667eea#764ba2)
  • Responsive design (mobile-friendly)
  • Mermaid.js support via CDN
  • Inline CSS (no external dependencies)
  • Clean typography and spacing

Template variables to replace:

  • {Report Title} - Main title
  • {Report Subtitle} - Subtitle
  • {Current DateTime} - Full timestamp
  • {Current Date} - Date only
  • {CONVERTED MARKDOWN CONTENT HERE} - HTML-converted content

Content Conversion Rules:

  1. Markdown to HTML: Convert markdown content to semantic HTML

    • # Header<h2>
    • ## Header<h3>
    • ### Header<h4>
    • **bold**<strong>
    • *italic*<em>
    • Lists → <ul> / <ol>
    • Code blocks → <pre><code>
    • Inline code → <code>
    • Tables → <table>
  2. Special Sections: Wrap important sections in cards

    • Decisions → .card div
    • Warnings → .card with warning style
    • Key points → .card
  3. Mermaid Diagrams: Detect markdown code blocks with mermaid language:

    ```mermaid
    graph TD
    A --> B
    ```
    

    Convert to:

    <div class="mermaid">
    graph TD
    A --> B
    </div>
    
  4. Timestamps: Use current datetime in header

Report Title Examples:

  • "Architecture Decisions"
  • "Project Status Dashboard"
  • "Quality Confidence Report"
  • "Component Documentation"

Step 6: Write latest.html

Use Write tool to create:

Location: .shipkit/communications/latest.html

Content: Complete HTML generated in Step 5

Example Write call:

Write tool:
  File: .shipkit/communications/latest.html
  Content: [Full HTML from Step 5]

Step 7: Confirm to User

After creating HTML, tell user:

✅ Visual HTML report created

📁 Location: .shipkit/communications/latest.html

📊 Report type: {Description}

{IF ARCHIVED}
🗄️  Previous report archived:
   .shipkit/communications/archive/{timestamp}-{description}.html
{END IF}

👉 Open latest.html in your browser to view the report.

Workspace Structure

This skill creates:

.shipkit/
  communications/
    latest.html                                    # Always current report
    archive/
      20251228-1430-architecture-decisions.html   # Previous reports
      20251228-1515-project-health.html
      20251228-1620-quality-checklist.html

Always one latest.html - Easy to find current report

Archive preserves history - Never lose previous visualizations


Completion Checklist

Copy and track:

  • Identified source content to visualize
  • Generated HTML with appropriate styling
  • Saved to appropriate location

Content Type Examples

1. Architecture Decisions

Source: .shipkit/architecture.json

Converts to: Visual timeline with cards for each decision, showing:

  • Decision date
  • What was decided
  • Rationale
  • Alternatives considered

Good for: Team alignment, onboarding, documentation


2. Project Status Dashboard

Source: All .shipkit/**/*.json files

Converts to: Comprehensive dashboard with:

  • Tech stack summary
  • Active specs (count + list)
  • Implementation plans (count + list)
  • Recent decisions
  • Component inventory

Good for: Stakeholder updates, weekly reviews


3. Quality Checklist

Source: .shipkit/codebase-index.json + specs

Converts to: Interactive checklist with:

  • All components listed
  • Acceptance criteria per spec
  • Implementation status
  • Quality gates

Good for: Pre-ship reviews, QA handoff


4. Component Documentation

Source: .shipkit/codebase-index.json

Converts to: Component catalog with:

  • Each component in a card
  • Props/interfaces
  • Usage examples
  • File locations

Good for: Developer reference, onboarding


Integration with Other Skills

Before shipkit-communications:

  • Any skill that creates .shipkit/ content
  • Examples:
    • /shipkit-engineering-definition → Visualize decisions
    • verify manually → Visualize checklist
    • document components manually → Visualize docs

After shipkit-communications:

  • Share HTML with team
  • Present in meetings
  • Continue current workflow

When to use:

  • Need visual format for presentation
  • Sharing with non-technical stakeholders
  • Want shareable artifact (email, Slack, etc.)
  • Prefer visual review over markdown

Context Files This Skill Reads

Depends on visualization type (see Step 2):

  • .shipkit/architecture.json - Architecture decisions
  • .shipkit/codebase-index.json - Components/routes
  • .shipkit/stack.json - Tech stack
  • .shipkit/specs/{todo,active,parked,shipped}/*.json - Specifications
  • .shipkit/plans/{todo,active,parked,shipped}/*.json - Implementation plans
  • Any .shipkit/**/*.json - Custom/comprehensive

Lazy loading: Only reads what's needed for chosen visualization


Context Files This Skill Writes

Run-scoped output (parallel-safe). communications/ is a transient artifact. When running under the orchestration engine, write under the run root (<runDir>/communications/…) per install/shared/references/run-artifacts.md; with no run context it stays at .shipkit/communications/… (back-compatible). Readers resolve the same base.

Creates/Updates:

  • .shipkit/communications/latest.html - OVERWRITE AND REPLACE
    • Write Strategy: OVERWRITE AND REPLACE
    • Behavior: Always contains most recent visualization
    • Why: "latest" means current - old versions go to archive

Archives:

  • .shipkit/communications/archive/{timestamp}-{description}.html - APPEND
    • Write Strategy: APPEND (create new file)
    • Behavior: Archives old latest.html before overwriting
    • Why: Preserve history, never lose previous visualizations

Never modifies source files: All .shipkit/*.json context files are read-only inputs


<!-- SECTION:after-completion -->

After Completion

Guardrails Check: Before moving to next task, verify:

  1. Persistence - Has important context been saved to .shipkit/?
  2. Prerequisites - Does the next action need a spec or plan first?
  3. Session length - Long session? Consider /shipkit-work-memory for continuity.

Natural capabilities (no skill needed): Implementation, debugging, testing, refactoring, code documentation.

Suggest skill when: User needs to make decisions, create persistence, or check project status.

<!-- /SECTION:after-completion --> <!-- SECTION:success-criteria -->

Success Criteria

Visualization is complete when:

  • User specified what to visualize
  • Source files read successfully
  • Old latest.html archived (if existed)
  • New latest.html created
  • HTML renders correctly in browser
  • Mermaid diagrams render (if present)
  • Styling matches ShipKit Lite purple theme
  • Mobile-responsive
  • User knows where to find latest.html
<!-- /SECTION:success-criteria -->

Common Scenarios

Scenario 1: Architecture Decisions for Team Review

User: "Visualize the architecture decisions"

Claude:
1. Ask confirmation (already clear)
2. Read: .shipkit/architecture.json
3. Check for latest.html (exists)
4. Archive old: archive/20251228-1430-architecture-decisions.html
5. Generate HTML with:
   - Timeline of decisions
   - Each decision as a card
   - Rationale highlighted
   - Alternatives in tables
6. Write: latest.html
7. Confirm: "✅ Architecture decisions visualized. Open .shipkit/communications/latest.html"

Scenario 2: Project Status for Stakeholders

User: "Create HTML report for weekly update"

Claude:
1. Ask: "What type of report?"
2. User: "Project status"
3. Glob: .shipkit/**/*.json
4. Read all markdown files
5. No existing latest.html (first time)
6. Generate comprehensive dashboard HTML
7. Write: latest.html
8. Confirm: "✅ Project status dashboard created"

Scenario 3: Quality Checklist Before Launch

User: "Show me quality checklist as HTML"

Claude:
1. Ask confirmation
2. Read: codebase-index.json + specs/{todo,active}/*.json
3. Check latest.html (exists)
4. Archive: archive/20251228-1620-quality-checklist.html
5. Generate checklist HTML with:
   - All acceptance criteria
   - Component verification
   - Quality gates
6. Write: latest.html
7. Confirm with next steps

Tips for Effective Visualizations

Choose the right type:

  • Architecture decisions → Best for team alignment
  • Project status → Best for stakeholder updates
  • Quality checklist → Best for pre-ship reviews
  • Component docs → Best for developer reference

When to regenerate:

  • After significant changes to source content
  • Before team meetings/presentations
  • Weekly for status updates
  • Before sharing with stakeholders

Archive benefits:

  • Compare how project evolved over time
  • Reference past visualizations
  • Track decision-making history
  • Never lose previous reports

Sharing tips:

  • latest.html is self-contained (inline CSS)
  • No external dependencies (except Mermaid CDN)
  • Works offline after initial load
  • Email-friendly (single file)
  • Mobile-responsive for phone viewing

Markdown to HTML Conversion Guide

Quick reference:

  • Headers: #<h2>, ##<h3>, ###<h4>
  • Emphasis: **bold**<strong>, *italic*<em>, `code`<code>
  • Lists: -<ul><li>, 1.<ol><li>
  • Code blocks with language support
  • Tables with header/body structure
  • Mermaid diagrams: ```mermaid<div class="mermaid">

Remember: This skill is about making ShipKit Lite content beautiful and shareable. Always read actual content, never generate placeholders. Archive old reports to preserve history.

What ships with it: 1 file

5.4 KB alongside SKILL.md

references/

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.