Systems Thinking Workflow Architect

Overview

Systems Thinking Workflow Architect helps an agent decompose complex problems and design reviewable process artifacts such as workflow maps, SOPs, decision trees, lead-generation processes, bottleneck analyses, and agent flow plans.

The skill is intended for planning, reasoning, and documentation. It does not execute workflows, make unsupervised high-stakes decisions, or replace legal, medical, financial, compliance, or other expert review.

When to use

Use this skill when a user asks for problem decomposition, systems thinking, workflow mapping, lead-generation process design, decision trees, SOP design, agent flow design, bottleneck analysis, or guidance on how an agent should think, decide, act, verify, and escalate.

Contents

• SKILL.md: agent-facing instructions, triggers, boundaries, workflow, output formats, quality checklist, and safety rules.
• README.md: this human-readable overview.
• agents/openai.yaml: OpenAI display metadata, invocation policy, and dependencies.
• references/style-guide.md: reusable patterns, templates, examples, and edge cases for workflow and systems design outputs.
• assets/: reserved for reusable templates or visual assets.
• scripts/example_helper.py: a safe deterministic helper for checking whether a drafted process artifact includes recommended review sections.

Usage examples

Map our inbound lead qualification workflow and identify bottlenecks.

Create a decision tree for routing support tickets to the right team, including escalation rules.

Design an SOP for how an AI agent should handle customer intake, verify missing information, and stop for human approval.

Decompose this operational problem using systems thinking and show the actors, feedback loops, constraints, and failure modes.

Package structure

systems-thinking-workflow-architect/
|-- SKILL.md
|-- README.md
|-- agents/
|   `-- openai.yaml
|-- assets/
|   `-- .gitkeep
|-- references/
|   `-- style-guide.md
`-- scripts/
    `-- example_helper.py

Notes for maintainers

• Keep trigger language specific and operational.
• Keep execution boundaries explicit.
• Put long examples, templates, and edge cases in references/style-guide.md rather than overloading SKILL.md.
• Add reusable files to assets/ only when they are safe to distribute.
• Keep scripts deterministic, non-destructive, and free of secrets.