stream-chain

stream-chain is an agent skill from the ruflo project for running multi-step workflows in which each agent's output flows into the next. Each step receives the complete output of the step before it, so the chain can perform complex data transformations and sequential processing through streamed JSON. The skill offers two modes. Custom chains, run with the run command, let you execute a sequence of prompts with full control over each step. Predefined pipelines, run with the pipeline command, use established workflows for common tasks. It is useful when a job breaks into a fixed order of stages, such as extracting facts, then summarizing them, then formatting the result, and when you want each stage's output to be inspected on its own. Because each step passes its whole output forward, a long chain can grow large and costly, so keep the steps focused and check the intermediate output when the result looks wrong. The skill is aimed at developers who build pipelines of agent calls and want a simple way to chain them. The skill passes Gen Agent Trust Hub, Socket and Snyk on skills.sh, and it has about 1,200 installs. Install it with npx skills add https://github.com/ruvnet/ruflo --skill stream-chain.…