stride

Orchestrate kanban task workflows across claim, explore, implement, review, and complete phases.

5|3|Updated Oct 19, 2025
One-click install
npx skills add https://github.com/cheezy/kanban --skill stride-cheezy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stride
Source: https://github.com/cheezy/kanban/tree/main/docs/multi-agent-instructions
Command: npx skills add https://github.com/cheezy/kanban --skill stride-cheezy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires stride-workflow, stride-claiming-tasks, stride-completing-tasks, stride-creating-tasks, stride-creating-goals, stride-enriching-tasks, stride-subagent-workflow.

What problem does it solve?

Stride provides a rigorous, automated workflow to manage AI-human collaboration on kanban tasks, ensuring every task goes through a defined lifecycle from claim to completion with enforced quality gates.

Core Features & Use Cases

  • Orchestrated lifecycle: claim → explore → implement → review → complete
  • Mandatory client-side hooks (before_doing, after_doing, before_review, after_review) and orchestrator-driven workflows
  • Sub-skills for enrichment, decomposition, exploration, planning, and review to structure complex projects
  • Telemetry-rich completions with structured workflow_steps and explorer/reviewer results
  • Guidance for safe task creation, API interaction, and error-handling in AI-assisted development

Quick Start

Prompt me to start a Stride task and I will execute the full orchestrated workflow from claim to complete.

Frequently Asked Questions about stride

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

FAQPage Schema
How do I automate AI-human task management on a kanban board?

You can automate AI-human task management by orchestrating claim, explore, implement, review, and complete phases with mandatory hooks. This enforces strict task field validation and structured telemetry for consistent governance across the entire task lifecycle.

What is an orchestrated workflow for AI-agent kanban tasks?

An orchestrated workflow for AI-agent kanban tasks is an automated lifecycle process moving items from claim to completion. It enforces quality gates through mandatory client-side hooks and sub-skills, ensuring consistent telemetry and error handling.

How do I enforce task field validation in an automated task workflow?

Task field validation is enforced by requiring strict fields like key_files, testing_strategy, and verification_steps during the workflow. This ensures every task meets defined quality gates before progressing through the completion phases.

Can I execute task workflows in both Claude Code and non-subagent environments?

Yes, the workflow operates across both Claude Code environments and non-subagent hosts. It maintains consistent governance, telemetry, and error handling regardless of the execution environment.

What are the limitations of using mandatory hooks for task lifecycle automation?

Mandatory hooks limit workflow flexibility by enforcing strict before_doing, after_doing, before_review, and after_review execution gates. This rigid structure ensures telemetry and quality control but requires strict adherence to the defined lifecycle phases.

How do I capture telemetry for task completion in an AI-agent orchestrator?

Telemetry is captured using a six-phase completion payload that records structured workflow_steps and explorer/reviewer results. This provides full lifecycle visibility for every completed kanban task.