beads

Name: beads
Availability: InStock
Author: Emmitt

Persist multi-session AI work context in a git-backed memory layer.

1|1|Updated Mar 9, 2026

Search Tags:#workflow #dependencies #beads #agents #memory #session

Ready to run

npx skills add https://github.com/EmmittJ/guild --skill beads-emmittj

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: beads
Source: https://github.com/EmmittJ/guild/tree/main/plugin/skills/setup/assets/skills/beads
Command: npx skills add https://github.com/EmmittJ/guild --skill beads-emmittj

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Beads provides a persistent, git-backed memory layer to capture and carry multi-session AI work context across sessions, preventing loss of decisions and context during conversation resets.

Core Features & Use Cases

Persistent memory: stores decisions, notes, and dependencies so work can resume after days or sessions.
Dependency tracking: supports blocks, discovered-from, and related relationships to model work relationships.
Session handoff and coordination: tracks agent states, hooks, and goals for collaborative work.
Audit-friendly memory: all changes are versioned and survive compaction.

Quick Start

Initialize a beads workspace and review the in_progress issues to resume work across sessions.

Frequently Asked Questions about beads

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

FAQPage Schema

How do I persist AI work context across multiple sessions?▼

Persistent AI work context is stored in a git-backed memory layer that captures decisions, notes, and dependencies, preventing loss of information during conversation resets. This allows work to resume seamlessly after days or sessions.

How does dependency tracking work for multi-session agent work?▼

Dependency tracking for multi-session agent work models relationships using blocks, discovered-from, and related links. This enforces dependency graphs and design notes, ensuring coordinated work survives across compaction.

What is the best way to hand off context between AI agents?▼

Cross-session handoff for AI agents is managed by tracking agent states, hooks, and goals. This coordination framework enforces acceptance criteria and persists work context so agents can resume collaborative tasks.

Does git-backed memory survive context window compaction?▼

Yes, git-backed memory survives context window compaction by versioning all changes to decisions and dependencies. This audit-friendly approach ensures no context is lost when the conversation is compacted.

How do I resume in-progress issues from a previous AI session?▼

To resume in-progress issues from a previous AI session, initialize a beads workspace to review the persisted state. The stored decisions and dependency graphs allow you to immediately continue where the prior session left off.