beads

Store AI agent decisions, notes, and dependencies in a git-backed memory.

418|40|Updated Nov 5, 2025
One-click install
npx skills add https://github.com/boshu2/agentops --skill beads-boshu2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: beads
Source: https://github.com/boshu2/agentops/tree/main/skills/beads
Command: npx skills add https://github.com/boshu2/agentops --skill beads-boshu2

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Beads provides a git-backed, persistent memory layer for AI agents, enabling long-running projects to survive conversation rollups and session boundaries by storing decisions, notes, and dependencies.

Core Features & Use Cases

  • Persistent memory: Git-backed storage for issues, design notes, and decision history across sessions.
  • Dependency-aware task graph: Supports blocks, related, parent-child, and discovered-from relationships to surface ready work automatically.
  • Session resilience and handoffs: Enables compaction recovery and cross-session handoffs so work can be resumed later with full context.
  • Scripting-free workflow: Command-based interactions to create, link, and manage work without external tooling.

Quick Start

Example commands:

  • bd create "Onboarding flow" -t epic -p 0
  • bd update onboarding-1 --status in_progress
  • bd ready

Frequently Asked Questions about beads

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

FAQPage Schema
How do I maintain AI agent memory across multiple sessions?

To maintain AI agent memory across multiple sessions, you need a git-backed persistent storage layer that captures decisions, notes, and dependencies, enabling work to survive conversation rollups and session boundaries.

What is a dependency-aware task graph for AI agents?

A dependency-aware task graph models relationships like blocks, related, parent-child, and discovered-from to track multi-session work, automatically surfacing ready tasks as blockers resolve.

How do I track complex dependencies and side quests in long-running AI projects?

You can track complex dependencies and side quests in long-running AI projects by applying a four dependency model—blocks, related, parent-child, and discovered-from—using command-based interactions to create, link, and update issues.

Do I need external scripting tools to manage AI session handoffs?

No, you do not need external scripting tools to manage AI session handoffs. You can use a scripting-free, command-based workflow to update statuses and surface ready work directly.

How does compaction recovery work for multi-session AI work?

Compaction recovery for multi-session AI work relies on git-backed persistent memory to restore full context, allowing compaction recovery and cross-session handoffs so work can be resumed later without losing decision history.