transformation-consultant

Transformation Consultant

An AI-powered business process transformation consultant plugin for Claude. Analyze process transcripts, generate BPMN 2.0 diagrams, and produce optimization recommendations with ROI analysis.

Overview

This plugin gives Claude deep expertise in business process transformation consulting. It can:

• Analyze process transcripts from interviews and walkthroughs to extract structured process information
• Generate BPMN 2.0 diagrams with APQC Level 4 activity consolidation for clean, standardized visualizations
• Produce optimization recommendations with automation opportunities, technology recommendations, ROI analysis, and implementation roadmaps

Installation

claude plugins add knowledge-work-plugins/transformation-consultant

Commands

Command	Description
/transformation-consultant:analyze-transcript	Analyze a business process transcript and extract structured process information
/transformation-consultant:generate-bpmn	Generate a BPMN 2.0 XML diagram from a process analysis
/transformation-consultant:optimize-process	Generate optimization recommendations with ROI analysis
/transformation-consultant:full-transformation	Run the complete pipeline: transcript analysis, BPMN generation, and optimization
/transformation-consultant:create-presentation	Generate an executive HTML slide presentation from recommendations

Skills

Skill	Description
transcript-analysis	Domain expertise in extracting structured process information from conversational transcripts
bpmn-generation	BPMN 2.0 diagram generation with APQC Level 4 activity consolidation and validation
process-optimization	Business process optimization consulting with prioritization frameworks and ROI analysis

Skills activate automatically when Claude detects relevant context in the conversation.

Example Workflows

Full Transformation (Recommended)

Run the complete pipeline on a process transcript:

/transformation-consultant:full-transformation data/sample-transcripts/ap-process.txt

This produces three outputs:

1. Process Analysis — Structured breakdown of steps, actors, decisions, and pain points
2. BPMN Diagram — Valid BPMN 2.0 XML ready for visualization
3. Optimization Recommendations — Prioritized improvements with ROI estimates

Step-by-Step Analysis

For more control, run each step individually:

# Step 1: Analyze the transcript
/transformation-consultant:analyze-transcript data/sample-transcripts/ap-process.txt

# Step 2: Generate BPMN from the analysis
/transformation-consultant:generate-bpmn outputs/analysis/ap-analysis.md

# Step 3: Get optimization recommendations
/transformation-consultant:optimize-process outputs/analysis/ap-analysis.md

Executive Presentation

After running the pipeline, generate an executive slide deck:

/transformation-consultant:create-presentation outputs/recommendations/ap-recommendations.md outputs/analysis/ap-analysis.md

This uses the frontend-slides skill to produce a polished HTML presentation. Requires the skill to be installed separately — see commands/create-presentation.md for setup.

With Business Context

Provide additional context for more targeted recommendations:

/transformation-consultant:optimize-process outputs/analysis/ap-analysis.md "Industry: Healthcare, Budget: $500K, Team size: 12 FTEs, Priority: reduce cycle time"

MCP Integration

Note: For connector setup and configuration details, see CONNECTORS.md.

This plugin can connect to external tools via MCP servers configured in .mcp.json:

• Slack — Share results with team channels
• Microsoft 365 — Read transcripts from SharePoint, save outputs to shared libraries

Sample Data

The data/sample-transcripts/ directory includes example transcripts for testing:

• ap-process.txt — Accounts Payable invoice processing walkthrough
• gl-month-end-close.txt — General Ledger month-end close process

Python Backup Mode

This project also includes a standalone Python implementation for programmatic execution. See SETUP.md for Python setup instructions.

# Run via Python CLI
python -m src.main data/sample-transcripts/ap-process.txt outputs/test

# Run via Python API
from src.main import run_full_transformation

result = run_full_transformation(
    transcript_path="data/sample-transcripts/ap-process.txt",
    output_dir="outputs/my-analysis"
)

Architecture

See ARCHITECTURE.md for detailed design documentation of the Python component architecture.

Contributing

Feedback and suggestions welcome! Fork the repo, edit the markdown/JSON files, and submit a PR.

License

MIT