baton

Store TLDR summaries and per-conversation state for long autonomous sessions.

Updated Dec 28, 2025
One-click install
npx skills add https://github.com/mgerasolo/habitarcade-poc --skill baton
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: baton
Source: https://github.com/mgerasolo/habitarcade-poc/tree/main/.claude-code/skills/baton
Command: npx skills add https://github.com/mgerasolo/habitarcade-poc --skill baton

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides robust context management for long autonomous sessions by maintaining TLDR summaries, per-conversation state, and efficient post-compaction recovery, ensuring no loss of critical details.

Core Features & Use Cases

  • TLDR summaries for quick recall of long sessions
  • Conversation tracking and switchable history across multiple conversations
  • Auto-save and post-compaction recovery to maintain continuity

Quick Start

Use /baton to initialize a new conversation, switch between conversations, and load TLDR summaries to resume work with preserved context.

Frequently Asked Questions about baton

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

FAQPage Schema
How do I preserve conversation context across sessions after compaction?

You can preserve conversation context across sessions after compaction by using TLDR summaries and per-conversation state tracking. This approach maintains critical details and enables rapid restoration of your workflow without losing prior progress.

What is the best way to manage context switching in multi-conversation workflows?

The best way to manage context switching in multi-conversation workflows is to maintain a switchable history with per-conversation state. This allows you to track and resume multiple contexts efficiently while keeping each session's details isolated.

How do I auto-save and recover context history in a local workspace?

You can auto-save and recover context history by integrating with the local .claude workspace. This mechanism handles loading, updating, and archiving context, ensuring safe post-compaction recovery and continuous workflow tracking.

How does a TLDR storage mechanism work for long autonomous sessions?

A TLDR storage mechanism works by generating concise summaries of long autonomous sessions. These summaries capture essential information, allowing you to quickly recall past interactions and maintain context retention without reviewing entire logs.

Can I use this context management approach for multi-conversation workflows without external dependencies?

Yes, you can use this context management approach for multi-conversation workflows without external dependencies. It relies entirely on the local .claude workspace to load, update, and archive context, functioning independently.