flow-next
NewZero-dependency planning + execution with .flow/ task tracking and Ralph autonomous mode (multi-model review gates). Worker subagent per task for context isolation. Includes 21 subagents, 22 commands, 26 skills.
Summary
flow/ task tracking and Ralph autonomous mode with multi-model review gates.
- It provides 21 subagents, 22 commands, and 26 skills to enable structured workflows, context isolation per task, and automated code review gates.
Install & Usage
mkdir -p .claude/agentsAdd the configuration to .claude/agents/flow-next.md
@flow-nextUse Cases
Usage Examples
/flow-next plan 'Add user authentication with JWT'
/flow-next execute --task-list .flow/tasks.json
Run Ralph autonomous mode to review all pending changes in the current branch.
Security Audits
Frequently Asked Questions
What is flow-next?
flow-next is a zero-dependency planning and execution framework for Claude Code that uses .flow/ task tracking and Ralph autonomous mode with multi-model review gates. It provides 21 subagents, 22 commands, and 26 skills to enable structured workflows, context isolation per task, and automated code review gates.
How to install flow-next?
To install flow-next: create the agents directory (mkdir -p .claude/agents), then add the config to .claude/agents/flow-next.md. Finally, @flow-next in Claude Code.
What is flow-next best for?
flow-next is a agent categorized under General. It is designed for: code-review, agent, claude-code, plugin, workflow, planning, execution. Created by Gordon Mickel.
What can I use flow-next for?
flow-next is useful for: Plan and execute a multi-step feature implementation with automatic task breakdown and progress tracking.; Run autonomous code review with multi-model review gates to catch issues before merging.; Isolate complex refactoring tasks into worker subagents to avoid context pollution.; Automate repetitive development workflows like testing, linting, and deployment steps.; Manage project-level task dependencies and track completion status across multiple files.; Set up a review pipeline that requires approval from different AI models before code is accepted..