mb-end

Compile session activity and spawn a subagent to crystallize insights.

35|6|Updated Jan 13, 2026
One-click install
npx skills add https://github.com/noontide-co/mainbranch --skill mb-end
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mb-end
Source: https://github.com/noontide-co/mainbranch/tree/main/.claude/skills/mb-end
Command: npx skills add https://github.com/noontide-co/mainbranch --skill mb-end

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Closes a session intentionally, surfacing what happened today and crystallizing insights when meaningful activity occurred. It serves as the bookend to /mb-start and ensures important decisions, research, and changes are captured and reviewed before you move on.

Core Features & Use Cases

  • Session wrap-up that surfaces today’s activity across research, decisions, and core changes
  • Crystallize step: spawn a dedicated subagent to surface tensions, gaps, and questions
  • Optional final thought capture and checkpoint save to preserve progress
  • End-of-day, end-of-research-batch, end-of-decision-sprint, or mid-work checkpoint scenarios

Quick Start

Run /mb-end at the end of a session to close intentionally and crystallize today's activity.

Frequently Asked Questions about mb-end

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

FAQPage Schema
How do I document a session checkpoint to capture decisions and research activity?

You should close a work session intentionally when meaningful activity occurs, such as at the end of a day, research batch, or decision sprint. Intentional session closure ensures important decisions and changes are captured and reviewed before you move on.

What is the best way to save context and surface tensions at the end of a work sprint?

The best way to save context and surface tensions is to use a crystallize subagent that reviews today's activity. This process surfaces gaps and questions, ensuring your decision trail and progress are preserved in a dedicated note.

How do I create a decision trail from today's work activity before ending a session?

You create a decision trail by running the mb-end command to compile today's activity across research and core changes. It spawns a subagent to crystallize insights, optionally captures final thoughts, and saves the progress to preserve context.

When should I intentionally close a work session instead of just stopping?

Intentionally close a work session when meaningful activity occurs, such as at end-of-day, end-of-batch, or mid-work checkpoints. This ensures important decisions, research, and changes are captured and reviewed before moving on.

Does ending a session automatically save my progress and final thoughts?

Ending a session does optionally save your progress and final thoughts. The skill enforces a sequence to summarize the session, capture optional final thoughts, spawn a crystallize subagent, and save a checkpoint note when meaningful activity exists.