builder

name	builder
description	Design and build AI-powered automation workflows. Use when creating integrations between systems, automating repetitive tasks, connecting APIs, or building workflows with Verdent, n8n, Make, or Zapier.
version	1.0.0

Builder

Note: Review PROFILE.md in this skill folder for user-specific automation priorities, platform preferences, and common use cases.

Master Briefing: Global brand voice at ~/.superskills/master-briefing.yaml applies automatically. Skill profile overrides when conflicts exist.

Design, build, and optimize AI-powered automation workflows using Verdent, n8n, and other platforms; connect systems via APIs to eliminate manual tasks.

Core Workflow

1. Requirements Definition

• Clarify automation objective (what manual work to eliminate)
• Identify trigger events and desired outcomes
• Map data sources, systems, APIs involved
• Confirm constraints (budget, timing, reliability)

2. Build

• Design workflow architecture (trigger → process → action → error handling)
• Select optimal platform (Verdent for AI, n8n for integrations)
• Build incrementally (happy path first, then edge cases)
• Implement comprehensive error handling and logging
• Test with real data in controlled environment

3. Deploy

• Deploy to production with monitoring
• Document setup, configuration, troubleshooting
• Provide usage instructions and runbook
• Handoff with training if needed

Quality Checklist

• Workflow processes 3+ test scenarios successfully
• Error handling covers failure modes
• Retry logic and fallback options implemented
• Logging sufficient for debugging
• Rate limits and API quotas respected
• Cost per execution acceptable
• Documentation includes architecture diagram
• Monitoring/alerting configured for critical failures

Platform Selection

Verdent (AI-Native IDE):

• Best for: AI agent workflows, complex multi-step AI tasks, MCP integrations
• Use when: Task requires AI decision-making, content creation, multiple MCP tools

n8n (Visual Workflow):

• Best for: API integrations, webhooks, scheduled automations, data transformation
• Use when: Connecting APIs, processing data, triggering on events

Make/Zapier (No-Code):

• Best for: Quick integrations, simple trigger-action patterns, prototyping
• Use when: Speed matters, pre-built connectors exist

Workflow Design Principles

1. Reliability:

• Error handling at every step (try/catch, retries, fallbacks)
• Graceful degradation (continue if non-critical fails)
• Alerting for critical failures
• Comprehensive logging

2. Efficiency:

• Minimize API calls (cache, batch operations)
• Parallel processing when independent
• Webhooks over polling
• Respect rate limits

3. Maintainability:

• Clear naming (workflows, variables, nodes)
• Visual organization (group nodes, comments)
• Documentation (architecture, setup, troubleshooting)
• Version control

4. Scalability:

• Handle variable data volumes (1 vs 1,000 items)
• Rate limit awareness, queue management
• Cost monitoring
• Resource optimization

Documentation Template

# [Workflow Name]

## Purpose
[What task automated? What problem solved?]

## Trigger
[What initiates: webhook, schedule, manual, file?]

## Process Flow
1. [Step 1: Action and tool/API]
2. [Step 2: Action and tool/API]
...

## Error Handling
- [Step X]: Retry 3x with backoff, fallback to [Y]
- [Alert]: Email on critical failure

## Cost
- API calls per run: [number]
- Est. cost: $[amount]/month at [volume]

## Troubleshooting
- **Error**: [Common error]
  **Solution**: [How to fix]

Common Use Cases

Content Distribution:

• Blog → Social posts (LinkedIn, Twitter, Instagram)
• Podcast → Blog + audiogram clips
• Video → Transcription + blog + social clips

Lead Management:

• Form → CRM entry + welcome email
• Newsletter signup → Tag + drip sequence
• Course enrollment → Access + onboarding

Research & Reporting:

• Weekly trend monitoring → Summary report
• Competitor tracking → Alert on key content
• Performance metrics → Auto dashboard updates

Avoid

• Fragile Pipeline: Linear, no error handling → Try/catch, retry with backoff, fallbacks
• API Waster: Redundant calls, polling → Cache, batch, use webhooks
• Black Box: No logging/docs → Log decisions, data transforms, errors; visual diagram
• Hard-Coded Chaos: Embed API keys, URLs → Use environment variables, centralize config

Escalate When

• Required API lacks docs or unreliable
• Cost per run exceeds budget
• Needs extensive custom coding (developer task)
• Data privacy/security concerns
• Workflow complexity suggests process redesign
• Multiple valid platforms with major trade-offs