workflow-designer

Design deterministic state machine workflows with guards, side effects, and Mermaid diagrams.

Updated May 9, 2026
One-click install
npx skills add https://github.com/GustavoGutierrez/engineering-skills --skill workflow-designer-gustavogutierrez
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: workflow-designer
Source: https://github.com/GustavoGutierrez/engineering-skills/tree/main/skills/workflow-designer
Command: npx skills add https://github.com/GustavoGutierrez/engineering-skills --skill workflow-designer-gustavogutierrez

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It solves the problem of turning complex, asynchronous business or agentic processes into precise workflow architectures that are deterministic, observable, and resilient to failures.

Core Features & Use Cases

  • State machine and orchestration design: Model workflows as FSMs with explicit states, transitions, guards, and side effects.
  • Saga and compensation planning: Define saga boundaries and compensating transactions for distributed rollback and recovery.
  • HITL checkpoint and observability design: Add human-in-the-loop decision points with timeouts plus metrics, traces, and alertable failure signals.
  • Resilience per step: Specify retry policies, timeouts, circuit breakers, idempotency mechanisms, and degraded-mode fallbacks.

Use it when you need an implementation-ready workflow blueprint for systems that span multiple services, require auditability/approvals, or must recover safely from worker crashes and transient downstream failures.

Quick Start

Use the workflow-designer skill to design a deterministic saga-based orchestration for an async “order fulfillment” process, including HITL approval checkpoints and per-step resilience and observability requirements.

Frequently Asked Questions about workflow-designer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design a saga pattern with compensating transactions for distributed rollback?

Saga pattern design models distributed workflows as deterministic state machines, defining explicit saga boundaries and compensating transactions to ensure safe rollback and recovery across multiple services. This approach guarantees durable state ownership and idempotency by design.

What is the best way to add human-in-the-loop checkpoints to an async workflow?

Human-in-the-loop checkpoints are added to async workflows by modeling explicit decision states within a state machine, complete with timeout policies and alertable failure signals to ensure the process degrades safely if approvals are delayed.

How do you model workflow state machines with guards and side effects?

Workflow state machines are modeled by defining explicit states and transitions, attaching guards to validate transition conditions, and specifying side effects for each step to ensure deterministic orchestration and observable execution.

How do I specify resilience and retry policies for each step of an orchestration pipeline?

Resilience per orchestration step is specified by defining explicit retry policies, timeouts, circuit breakers, and degraded-mode fallbacks, ensuring the workflow recovers safely from worker crashes and transient downstream failures.

Does workflow design as a state machine support observability and auditability?

State machine workflow design supports observability and auditability by attaching metrics, traces, and alertable failure signals to each transition, ensuring explicit state ownership and comprehensive sad-path coverage for distributed systems.

When should I not use deterministic state machines for async business processes?

Deterministic state machines are less suitable for async business processes when workflows are highly ad-hoc, lack explicit state boundaries, or require no auditability, compensations, or resilience planning for worker crashes and transient failures.