distributed-workflows

Coordinate long-running distributed workflows with saga-based orchestration and compensation.

Updated Mar 29, 2026

Search Tags:#timeouts #compensation #distributed-systems #workflows #saga #idempotence #orchestrate

Ready to run

npx skills add https://github.com/marquesfelip/agents-and-skills --skill distributed-workflows-marquesfelip

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: distributed-workflows
Source: https://github.com/marquesfelip/agents-and-skills/tree/main/skills/distributed-workflows
Command: npx skills add https://github.com/marquesfelip/agents-and-skills --skill distributed-workflows-marquesfelip

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinate long-running distributed workflows across multiple services and external systems.

Core Features & Use Cases

Stateful saga orchestration with options for choreography or orchestration, enabling reliable multi-service coordination
Step-level timeouts, compensation, and idempotent restarts for reliability and auditability
Built-in stuck-workflow detection, persistence, and expiration to support fault tolerance and compliance

Quick Start

Configure and start a new workflow instance using the provided workflow definition and a unique correlation ID.

Frequently Asked Questions about distributed-workflows

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

FAQPage Schema

How do I coordinate long-running distributed workflows across multiple services?▼

Saga orchestration manages distributed workflows by maintaining a workflow state model that tracks each step, enforcing timeouts, and executing reverse-compensation to undo failed operations. This ensures reliable multi-service coordination with full auditability.

What is the best way to handle compensation and rollback in distributed systems?▼

Reverse-compensation handles rollback in distributed systems by defining specific undo actions for each completed step when a later step fails. This saga-based approach ensures your multi-service workflow can cleanly revert partial progress without leaving inconsistent state.

How do I retry stuck workflows with idempotent startup?▼

You retry stuck workflows using idempotent startup combined with a background stuck-workflow detector that identifies and attempts to recover stalled instances. This ensures safe restarts without duplicating side effects across your distributed services.

Can I use saga orchestration for SaaS provisioning and payment workflows?▼

Yes, saga orchestration is explicitly suitable for SaaS provisioning and payment workflows. It coordinates long-running processes across external systems by providing step-level timeout enforcement, persistent state tracking, and expiration handling for fault tolerance.

How do you enforce timeouts on individual steps in a distributed workflow?▼

Step-level timeout enforcement in a distributed workflow triggers expiration and recovery actions when a specific service step exceeds its allotted time. This prevents indefinite hanging and allows the orchestration engine to initiate compensation or retry logic.

When do I need saga orchestration instead of standard service calls?▼

You need saga orchestration when a business process spans multiple services and requires compensation, timeouts, retries, or audit trails. Standard service calls lack the state model and reverse-compensation needed to reliably reverse partial failures in long-running distributed systems.