Analyze local repository Skill

Agent Relationship Flow

The builder ecosystem follows a clear delegation chain:

┌─────────────────────────────┐
│  builder-reverse-engineer   │ ← FIRST: Analyze repo, extract patterns
│  (moai-builder/)            │
└──────────────┬──────────────┘
               │
               ▼ identifies automation opportunities
┌─────────────────────────────┐
│  builder-workflow-designer  │ ← THEN: Creates SPECs, diagrams
│  (moai-builder/)            │
└──────────────┬──────────────┘
               │
               ▼ delegates script generation
┌─────────────────────────────┐
│  builder-workflow           │ ← FINALLY: Creates UV scripts
│  (moai-builder/)            │
└─────────────────────────────┘

Workflow Stages

Stage 1: Repository Analysis (builder-reverse-engineer)

Scan repository structure and file organization
Identify technology stack and dependencies
Extract design patterns and conventions
Discover automation opportunities
Output: TOON-formatted analysis report

Stage 2: SPEC & Diagram Creation (builder-workflow-designer)

Parse analysis report
Create EARS-style SPEC documents
Generate workflow diagrams
Design agent coordination patterns
Output: SPEC documents, Mermaid diagrams

Stage 3: Implementation (builder-workflow, builder-skill, builder-agent)

Generate UV CLI scripts from patterns
Create skill structures with SKILL.md
Build agent blueprints with proper frontmatter
Output: Production-ready code and configurations

Available Commands

The builder ecosystem exposes these commands:

`/builder:reverse-engineer [repo-path]`

Analyze repository for patterns and automation opportunities.

# Analyze local repository
/builder:reverse-engineer /path/to/repo

# Analyze GitHub repository
/builder:reverse-engineer https://github.com/owner/repo

`/builder:workflow-designer [analysis-output]`

Create SPECs and diagrams from analysis.

# Design workflow from analysis
/builder:workflow-designer .moai/reports/analysis-report.md

`/builder:generate-script [skill-name] [script-name]`

Generate UV CLI script for existing skill.

# Generate analysis script
/builder:generate-script moai-foundation-core analyze-compliance

`/builder:generate-skill [skill-name] [description]`

Generate new skill with SKILL.md structure.

# Generate new skill
/builder:generate-skill moai-connector-slack "Slack integration with MCP"

`/builder:generate-workflow [workflow-name]`

Generate complete workflow (skill + agents + scripts).

# Generate complete workflow
/builder:generate-workflow api-testing-automation

Integration with builder-skill-uvscript

This skill works alongside builder-skill-uvscript which contains 23 production UV scripts:

Script Categories in builder-skill-uvscript:

System Analysis: 12 scripts (analyze_, monitor_, optimize_*)
Code Generation: 4 scripts (generate_agent, generate_skill, generate_command, scaffold_test)
Development Tools: 2 scripts (debug_code, analyze_performance)
Documentation: 2 scripts (lint_docs, validate_diagrams)
BaaS Integration: 2 scripts (select_provider, migrate_provider)
Template Tools: 1 script (generate_template)

Relationship:

builder-workflow (this skill) - Orchestration patterns and agent coordination
builder-skill-uvscript - Reference implementations and production scripts

Workflow Patterns

Pattern 1: Repository-to-Skill Pipeline

pipeline:
  input: Repository URL or path
  stages:
    - builder-reverse-engineer → Analysis Report
    - builder-workflow-designer → SPEC + Diagrams
    - builder-skill → Skill Structure
    - builder-workflow → UV Scripts
  output: Complete skill with scripts

Pattern 2: Quick Script Generation

pipeline:
  input: Skill name + Script requirements
  stages:
    - builder-workflow → UV Script
  output: Single production script

Pattern 3: Agent Blueprint Creation

pipeline:
  input: Agent requirements
  stages:
    - builder-reverse-engineer → Pattern analysis
    - builder-agent → Agent YAML with frontmatter
  output: Agent definition file

Quality Standards

All generated artifacts follow:

TRUST 5 Principles:

Testable: Includes test patterns
Readable: Clear documentation
Understandable: Self-documenting code
Secure: No hardcoded secrets
Traceable: Version history

IndieDevDan 13 Rules (for UV scripts):

200-300 line target
PEP 723 dependencies
Zero shared imports
Dual output modes
Project-root detection

MoAI Standards:

YAML frontmatter
Color coding (yellow = custom)
Agent hierarchy compliance
Command naming conventions

Related Skills

| Skill | Purpose | |-------|---------| | builder-skill-uvscript | 23 reference UV scripts | | moai-foundation-core | Execution rules, TRUST 5 | | moai-lang-unified | Multi-language patterns | | moai-library-toon | TOON format specification |

Version: 1.0.0 Status: Active Last Updated: 2025-12-01 Color: Yellow (Custom Superdisco)

Agent Skills: Analyze local repository

Install this agent skill to your local

Skill Files