ticks

Track and preserve task state for AI agents across sessions and worktrees.

6|1|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/pengelbrecht/ticks --skill ticks
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ticks
Source: https://github.com/pengelbrecht/ticks/tree/main/skills/ticks
Command: npx skills add https://github.com/pengelbrecht/ticks --skill ticks

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Tracks and preserves task state for AI agents across sessions, enabling persistent memory and coordinated work within a repository.

Core Features & Use Cases

  • Persistent memory: issues live in your repo as git-tracked JSON files with fast local querying and offline access.
  • Agent-driven workflows: run agents on epics, create ticks from SPEC.md, and manage collaborative tasks with owner scoping.
  • Rich collaboration: multiplayer-first design with git worktrees to support teams of agents and humans.

Quick Start

Initialize the .tick directory, create SPEC.md if needed, and start the first tick with tk commands.

Frequently Asked Questions about ticks

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

FAQPage Schema
How do I track AI agent tasks across git worktrees?

You can track AI agent tasks across git worktrees by storing issues as git-tracked JSON files, which preserves task state and context across sessions for persistent memory and offline access.

How does task state persistence work for AI agents in a repository?

Task state persistence for AI agents works by saving issues as local JSON files within your repository, enabling fast local querying and coordinated work without losing context between sessions.

How do I create tasks from a SPEC.md file using a CLI?

You can create tasks from a SPEC.md file by initializing the .tick directory and using the tk CLI commands to generate SPEC-driven ticks for your agent-driven workflow.

Can I coordinate multiple AI agents on the same epic?

Yes, you can coordinate multiple AI agents on the same epic using the multiplayer-first design that supports teams of agents and humans through git worktrees and owner scoping.

What is the best way to manage human approvals in AI agent workflows?

The best way to manage human approvals in AI agent workflows is using a task tracker that defines a robust workflow loaded from SKILL.md, ensuring safe autonomous use with human checkpoints.

Does this issue tracker require an external database for task tracking?

No, it does not require an external database because issues live directly in your repository as git-tracked JSON files, providing fast local querying and full offline access.