flow

Orchestrate end-to-end feature lifecycles from creation to merge in Claude harness.

79|9|Updated Dec 9, 2025
One-click install
npx skills add https://github.com/panayiotism/claude-harness --skill flow-panayiotism
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: flow
Source: https://github.com/panayiotism/claude-harness/tree/main/skills/flow
Command: npx skills add https://github.com/panayiotism/claude-harness --skill flow-panayiotism

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Flow orchestrates an end-to-end feature lifecycle inside Claude harness, coordinating creation, planning, implementation, verification, checkpoint, and merging across teams. It solves the complexity of coordinating multi-phase software delivery, enabling deterministic progress through isolated contexts and memory-aware workflows.

Core Features & Use Cases

  • Unified lifecycle management from creation to merge, including planning, ATDD, and autonomous batch flow.
  • Optional Agent Teams support for parallel ATDD roles (tester, implementer, reviewer) to speed up delivery.
  • Context isolation via subagents to preserve clean state between features and ensure deterministic results across iterations.

Quick Start

Start a full lifecycle by running the flow command with your feature description.

Frequently Asked Questions about flow

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

FAQPage Schema
How do I automate an end-to-end feature workflow from planning to automated PR management?

Autonomous batch processing allows multiple features to progress through the lifecycle simultaneously. Flow coordinates context isolation via subagents, preserving clean state between features to ensure deterministic results across iterations.

How does ATDD work with agent teams in an automated software delivery pipeline?

Context isolation via subagents preserves clean state between features, ensuring deterministic results across iterations. This prevents context bleed and maintains structured memory updates throughout the feature lifecycle.

Do I need any external dependencies to run an autonomous feature lifecycle workflow?

Context isolation via subagents preserves clean state between features, ensuring deterministic results across iterations. This prevents context bleed and maintains structured memory updates throughout the feature lifecycle.

What is the best way to coordinate multi-phase software delivery for development teams?

Context isolation via subagents preserves clean state between features, ensuring deterministic results across iterations. This prevents context bleed and maintains structured memory updates throughout the feature lifecycle.

Why use checkpoint merging instead of manual git operations for feature integration?

Context isolation via subagents preserves clean state between features, ensuring deterministic results across iterations. This prevents context bleed and maintains structured memory updates throughout the feature lifecycle.