ba

Track AI-session tasks with file-based CLI commands in .ba/ directories.

3|Updated Jan 7, 2026
One-click install
npx skills add https://github.com/open-horizon-labs/bottle --skill ba-open-horizon-labs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ba
Source: https://github.com/open-horizon-labs/bottle/tree/main/codex-skill/ba
Command: npx skills add https://github.com/open-horizon-labs/bottle --skill ba-open-horizon-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

ba provides simple, file-based task tracking for AI sessions, eliminating scattered notes and ad-hoc progress updates in long-running experiments.

Core Features & Use Cases

  • Lightweight tracking: stores tasks in .ba/ without servers or databases.
  • CLI-driven workflow: create, claim, update, and complete tasks using the ba commands.
  • Collaborative-friendly: supports shared workspaces for teams collaborating on AI sessions.

Use Case: In an AI development session, you create tasks for planning, coding, evaluation, and review; team members claim tasks, monitor progress, and complete tasks to close work.

Quick Start

Install the ba CLI if not installed, then initialize tracking for your project and manage tasks with ba status, ba claim <id>, ba finish <id>, and ba create "<title>".

Frequently Asked Questions about ba

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

FAQPage Schema
How do I track AI session tasks with local files instead of a database?

Track AI session tasks with local files using a lightweight, file-based tracker that stores all task statuses in a local .ba/ directory without requiring external servers or databases. It provides simple CLI commands to create, claim, update, and complete tasks for individuals or small teams.

What's the best way to manage task status in local projects for small teams?

Manage task status in local projects by initializing a shared workspace with a file-based tracker. Small teams can collaborate on AI sessions by claiming, updating, and completing tasks via CLI commands, ensuring visible progress tracking without relying on external project management software.

Can I use a CLI-driven workflow to create and assign tasks during AI development?

Yes, you can use a CLI-driven workflow to create and assign tasks during AI development. Use commands like ba create to add planning, coding, and evaluation tasks, then let team members claim specific task IDs to coordinate work within the shared local project.

Do I need an external server to maintain task tracking for long-running experiments?

No, you do not need an external server to maintain task tracking for long-running experiments. The tracker relies entirely on local persistence within the .ba/ directory, eliminating scattered notes and ad-hoc progress updates without database dependencies.

How does file-based task tracking work for collaborative AI sessions?

File-based task tracking works for collaborative AI sessions by storing tasks in a shared .ba/ directory within the project. Team members use CLI commands to claim, monitor, and finish tasks, keeping work status visible across the shared workspace.

Why use a lightweight task tracker for AI session workflows?

Use a lightweight task tracker for AI session workflows to eliminate scattered notes and ad-hoc progress updates. It provides visible task status and simple CLI commands in local projects, satisfying the need for efficient tracking without the overhead of complex project management tools.