session-brain

Persist per-repo coding context in brain.json and brain.md files.

10|1|Updated Dec 24, 2025
One-click install
npx skills add https://github.com/zenbase-ai/code-voyager --skill session-brain
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-brain
Source: https://github.com/zenbase-ai/code-voyager/tree/main/.claude/skills/session-brain
Command: npx skills add https://github.com/zenbase-ai/code-voyager --skill session-brain

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides persistent, per-repo memory of goals, decisions, and progress to reduce context switching and rework.

Core Features & Use Cases

  • Tracks current_goal, current_plan, open_questions, and risks
  • Exposes brain.json for programmatic querying and brain.md for human-friendly summaries
  • Automatically updates via hooks to stay in sync with your work

Quick Start

Ask Claude Code to resume where you left off, or query what's next. Inspect brain.json with jq for structured data.

Frequently Asked Questions about session-brain

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

FAQPage Schema
How do I maintain coding context across multiple sessions in the same repository?

Session-brain provides persistent per-repo working memory that automatically tracks your goals, decisions, and progress across sessions. It stores canonical data in brain.json and human-readable summaries in brain.md under .claude/voyager/, so you can resume work without losing context.

What information does session-brain track and store?

Session-brain tracks current goals, plans, open questions, and risks specific to your coding work. It exposes both brain.json for programmatic queries and brain.md for human-friendly summaries, updated automatically via hooks to stay synchronized with your work.

How does session-brain keep memory synchronized with my work?

Session-brain implements automatic state persistence using hooks that update brain.json and brain.md whenever you make progress. This ensures your repository's memory stays current without manual intervention, reducing rework and context-switching overhead.

Can I query session-brain's memory programmatically?

Yes, session-brain exposes brain.json for structured, programmatic access to your session data. You can use tools like jq to inspect and query goals, decisions, progress, and open questions directly from the JSON file.

Do I need to manually update my coding progress notes?

No, session-brain automatically updates via hooks integrated into your workflow. Your goals, decisions, and progress sync continuously, so you focus on coding while the Skill maintains your repository's memory.

What happens to my session history when I switch repositories?

Session-brain maintains per-repository memory, so switching repos keeps each project's context isolated. Stored under .claude/voyager/ in each repo, your session episodes and memory persist independently for each codebase.