Stream outputs between sequential steps in multi-agent workflows.
npx skills add https://github.com/pacphi/ampel --skill stream-chain-pacphi
Please help me install this Agent Skill. Skill: stream-chain Source: https://github.com/pacphi/ampel/tree/main/.claude/skills/stream-chain Command: npx skills add https://github.com/pacphi/ampel --skill stream-chain-pacphi
Stream-Chain enables the construction of streaming, multi-agent workflows where each agent's output becomes the input for the next step. It supports data transformations and sequential processing, reducing manual handoffs and enabling scalable, reproducible pipelines.
Use Case Example: Build a pipeline that analyzes a codebase, extracts insights, and then generates an actionable improvement plan, with each stage feeding the next.
claude-flow stream-chain run "Analyze codebase structure" "Identify improvement areas" "Generate action plan"
claude-flow stream-chain pipeline analysis
High-intent search queries and answers about installing and using this skill.
Stream-chain orchestrates multi-agent workflows by streaming each agent's output as input to the next step. Define custom chains with sequential prompts or use predefined pipelines for software development tasks like analysis, refactoring, and testing, with cross-step context propagation and per-step streaming data flow.
Yes. Stream-chain's run mode lets you define deterministic prompt sequences with full control over data flow between steps. Each step receives complete output from the previous step, enabling rich streaming coordination across agents without manual handoffs.
Stream-chain includes battle-tested predefined workflows accessed via the pipeline mode for common software development tasks: codebase analysis, improvement identification, refactoring, testing, and optimization. Each pipeline stages multi-agent processing with automatic data propagation.
Stream-chain eliminates manual data transfer by automatically streaming output from each agent directly to the next step. Sequential processing with complete context propagation across steps enables scalable, reproducible pipelines without intermediate manual intervention.
Stream-chain supports verbose and debug modes for monitoring, plus timeout configuration to control step execution duration. These options provide visibility into streaming data flow and help diagnose issues in multi-agent pipeline execution.
Yes. Stream-chain is designed for this pattern: analyze codebase structure in the first step, extract insights in the second, and generate actionable improvement plans in the third. Each stage feeds the complete output to the next, creating coherent multi-agent analysis.
yashurathodOrchestrate multi-agent stream workflows by passing outputs between steps.
MinhThang1009Aggregate agent outputs into a deduplicated FINDINGS_REPORT.md sorted by severity.
NourcineAbOrchestrate sequential AI workflows by chaining prompts and transforming outputs across processing steps.
Human-Frontier-Labs-IncStream each step's output into the next in multi-agent workflows via claude-flow CLI.
