playbooks

home / skills / andrelandgraf / fullstackrecipes / workflow-setup

workflow-setup skill

/.agents/skills/workflow-setup

This skill guides you through installing and configuring the Workflow Development Kit for resumable AI agent workflows with step-level persistence.

npx playbooks add skill andrelandgraf/fullstackrecipes --skill workflow-setup

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
565 B
---
name: workflow-setup
description: Install and configure the Workflow Development Kit for resumable, durable AI agent workflows with step-level persistence, stream resumption, and agent orchestration.
---

# Workflow Development Kit Setup

To set up Workflow Development Kit Setup, refer to the fullstackrecipes MCP server resource:

**Resource URI:** `recipe://fullstackrecipes.com/workflow-setup`

If the MCP server is not configured, fetch the recipe directly:

```bash
curl -H "Accept: text/plain" https://fullstackrecipes.com/api/recipes/workflow-setup
```

Overview

This skill installs and configures the Workflow Development Kit to build resumable, durable AI agent workflows with step-level persistence, stream resumption, and agent orchestration. It provides practical setup steps and integration guidance for full-stack web AI apps, including fetching an official recipe from the fullstackrecipes MCP server or directly via curl. The goal is to get a production-ready workflow runtime running quickly and reliably.

How this skill works

The skill guides you through fetching the workflow-setup recipe either from an MCP server endpoint or via a direct HTTP request. It walks through installing the kit, configuring persistence and resume settings, and wiring agent orchestration so workflows can survive restarts and resume from the last completed step. It also outlines recommended configuration for TypeScript projects and server integration patterns.

When to use it

  • When you need durable, resumable AI workflows that persist progress across restarts or failures.
  • When orchestrating multiple agents or long-running steps that must resume mid-stream.
  • When building full-stack AI features that require reliable state and auditability.
  • When you want production-ready patterns and recipes tailored for TypeScript stacks.

Best practices

  • Enable step-level persistence and choose a reliable backing store (database, object store) early in setup.
  • Use idempotent step implementations so resumed executions are safe to replay.
  • Configure stream resumption timeouts and retention policies to balance cost and reliability.
  • Integrate clear observability (logs, traces, metrics) for workflow lifecycle events.
  • Keep orchestration logic declarative and isolate side effects to defined steps.

Example use cases

  • Long-running data enrichment pipelines that continue after process crashes or deployments.
  • User-facing multi-step flows (onboarding, approvals) that must preserve progress across sessions.
  • Agent-to-agent workflows where one agent triggers another and state must be tracked.
  • Batch processing jobs that require checkpointing and safe restart semantics.

FAQ

How do I fetch the recipe if the MCP server is not available?

Use the direct HTTP endpoint with curl: curl -H "Accept: text/plain" https://fullstackrecipes.com/api/recipes/workflow-setup.

What language and stack does this setup target?

This skill is geared toward TypeScript full-stack projects and includes patterns and configuration suited for Node.js environments.