Chain streaming outputs from one step as full context into the next.
npx skills add https://github.com/monktui/git-review --skill stream-chain-monktui
Please help me install this Agent Skill. Skill: stream-chain Source: https://github.com/monktui/git-review/tree/main/.claude/skills/stream-chain Command: npx skills add https://github.com/monktui/git-review --skill stream-chain-monktui
Stream-Chain helps you coordinate multi-step AI work where each agent’s output must become the direct input to the next step, so you can reliably transform ideas into sequential results instead of restarting or manually stitching outputs.
run mode for bespoke workflows or ready-made pipeline modes for common tasks like analysis, refactoring, testing, and optimization.Run a chained workflow for code changes by executing: claude-flow stream-chain run "Analyze the current codebase and identify key issues" "Propose specific improvements with a prioritized plan" "Generate implementation guidance plus tests for the top recommendations."
High-intent search queries and answers about installing and using this skill.
To chain multi-agent outputs into a sequential pipeline, use a custom run mode that passes each step's full streaming output as direct context into the next step. This ensures intermediate results are preserved and reused without manual stitching.
Yes, you can run predefined workflows for code refactoring and test generation. The system includes ready-made pipeline modes for common development tasks like analysis, refactoring, testing, and optimization, alongside flexible custom chains.
You orchestrate multi-step code analysis workflows by executing a chained command with sequential prompts, such as analyzing the codebase, proposing prioritized improvements, and generating implementation guidance with tests for the top recommendations.
Pipeline orchestration supports per-step timeouts and debug execution to improve observability. These operational controls help you monitor sequential workflows and reduce iteration time during complex multi-agent tasks.
The best way to maintain context across multi-agent data transformation steps is sequential chaining, where each agent receives the full previous output. This mechanism guarantees deeper continuity without restarting the workflow.
xotongStream each agent's output into the next step for multi-agent workflows.
moogieonAutomate merge, test, review, and deployment workflows with Git and CI/CD.
Dorpeer95Stream outputs between sequential steps in multi-agent workflow pipelines.
EthansuttorCoordinates multi-agent workflows by decomposing objectives into executable subtasks and unifying results.