Process Mapper — Business

Q: Can I use this for a one-off project?

No, it is designed for repeatable business workflows with discrete stages, not one-off projects.

Process Mapper documents end-to-end business processes in BPMN-style notation, measures cycle times by stage, and identifies bottlenecks using deterministic rules. It produces a process map, a ranked bottleneck list with severity and root-cause hypotheses, and a cycle-time analysis including value-add ratio and Little's-Law throughput.

business-operations claude-code

📂 View skill on GitHub ↗ 🚀 How to use

✨ What it does

Generates ASCII swim-lane diagrams with lanes by owner and stage annotations.
Computes total P50/P90 cycle time, value-add ratio, wait %, rework %, and Little's-Law throughput.
Applies three deterministic bottleneck detection rules with severity and root-cause hypotheses.
Adjusts detection thresholds by industry profile (saas, services, manufacturing, healthcare).
Provides Lean/Six Sigma canon references and anti-pattern guidance.
Includes a forcing-question library to ensure data quality before mapping.

🎯 When to use it

Documenting a new business process such as procurement intake, employee onboarding, or claims adjudication.
An existing process is 'too slow' but nobody can name the bottleneck.
Cycle time is measured but value-add ratio is not, so process health is unclear.
Cross-functional handoffs are dropping work and root cause is unclear.

🚀 How to use

Trigger: /cs:grill-bizops

Trigger by asking to document or analyze a business process, e.g., 'Map our procurement process and find bottlenecks.' Provide stage-level data as a JSON file with name, owner, type (value-add/wait/rework), and P50/P90 durations. The skill runs three scripts in sequence: process_documenter.py, bottleneck_detector.py, and cycle_time_analyzer.py. Example prompts:

Map the employee onboarding process and identify bottlenecks.
Analyze cycle time for our claims adjudication process using the attached JSON.

📄 Output: A process map (ASCII swim-lane diagram), a ranked bottleneck list, and a cycle-time analysis report.

📦 Add this skill to Claude Code

# 1. Get the skills repo
git clone --depth 1 https://github.com/alirezarezvani/claude-skills /tmp/claude-skills

# 2. Copy this skill into your project (or ~/.claude/skills for all projects)
mkdir -p .claude/skills
cp -r /tmp/claude-skills/business-operations/skills/process-mapper .claude/skills/process-mapper

Skill source: business-operations/skills/process-mapper/SKILL.md

⚠️ Good to know

Requires stage-level cycle-time data (even rough estimates); if unavailable, the process must be instrumented first.

❓ FAQ

What data do I need to provide?

You need a JSON file with one entry per stage, including name, owner, type (value-add, wait, or rework), and P50/P90 durations in minutes.

How does the skill determine if a process is healthy?

It uses the value-add ratio: >25% is healthy, 10-25% is typical, and <10% is waste-heavy.

Can I use this for a one-off project?