stream-chain

Stream each step's output into the next for multi-agent workflows.

1|Updated Jul 26, 2025
One-click install
npx skills add https://github.com/jcolano/claude-flow --skill stream-chain-jcolano
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stream-chain
Source: https://github.com/jcolano/claude-flow/tree/main/.claude/skills/stream-chain
Command: npx skills add https://github.com/jcolano/claude-flow --skill stream-chain-jcolano

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Executing complex, multi-step AI tasks often requires manual intervention to pass outputs from one agent to the next. This Skill automates that process, creating seamless pipelines for data transformation and sequential processing.

Core Features & Use Cases

  • Custom Chains: Define and execute custom sequences of prompts, where each step's output feeds into the next.
  • Predefined Pipelines: Utilize battle-tested workflows for common tasks like analysis, refactoring, testing, and optimization.
  • Context Preservation: Ensures the full output history flows sequentially through all steps, maintaining context across the entire workflow.
  • Use Case: Automate a full development cycle: analyze codebase, identify improvement areas, generate a refactoring plan, apply refactoring, and then create unit tests, all in one continuous flow.

Quick Start

Run a custom chain to write a user authentication function, add input validation, and then create unit tests for it. claude-flow stream-chain run "Write a user authentication function" "Add input validation and error handling" "Create unit tests with edge cases"

Frequently Asked Questions about stream-chain

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

FAQPage Schema
How do I chain multiple AI agents together to process data sequentially?

Chain AI outputs by defining custom sequences where each step's output feeds into the next, automating multi-agent workflows for data transformation and sequential processing without manual intervention between steps.

Can I automate a multi-step workflow like code analysis, refactoring, and testing in one continuous flow?

Yes, stream-chain orchestrates end-to-end workflows by piping each agent's full output into the subsequent step, enabling automated pipelines for code analysis, security audits, and testing sequences with context preservation throughout.

What's the difference between custom chains and predefined pipelines?

Custom chains let you define arbitrary prompt sequences tailored to your task, while predefined pipelines offer battle-tested workflows for common operations like analysis, refactoring, testing, and optimization with built-in best practices.

How does context flow through multi-step AI workflows?

Stream-chain preserves the full output history sequentially through all steps, ensuring each downstream agent receives complete context from prior steps rather than isolated inputs, maintaining coherence across the entire workflow.

Can I control execution with timeouts and debug output?

Yes, stream-chain exposes configurable options including --verbose for detailed logging, --timeout for execution limits, and --debug for troubleshooting, giving you fine-grained control over multi-agent workflow execution.

When should I use workflow automation instead of running AI steps manually?

Use stream-chain when executing complex, multi-step AI tasks that require passing outputs between agents; automation eliminates manual intervention, reduces errors, and enables consistent, repeatable pipelines for data transformation and sequential processing.