beads

Track task dependencies and persistent memory across AI agent sessions.

24|4|Updated Dec 26, 2025
One-click install
npx skills add https://github.com/NguyenSiTrung/Conductor-Beads --skill beads-nguyensitrung
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: beads
Source: https://github.com/NguyenSiTrung/Conductor-Beads/tree/main/.claude/skills/beads
Command: npx skills add https://github.com/NguyenSiTrung/Conductor-Beads --skill beads-nguyensitrung

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Beads provides persistent memory for AI agents across conversation compaction, enabling long-running projects to resume with full context. It introduces a graph-based issue tracker (bd) and stores state in .beads, allowing cross-session memory, dependencies, and audit trails.

Core Features & Use Cases

  • Dependency graphs: link tasks and track readiness
  • Cross-session persistence: state survives context resets
  • Audit trails: notes, decisions, and blockers persist
  • Beads integration with Conductor: tracks map to epics and tasks
  • Git-backed: versioned state via .beads/*.jsonl or DB
  • auto-sync and workflows with bd commands

Quick Start

Install bd, run bd init to create Beads memory, then use bd ready to fetch tasks and bd done to complete.

Frequently Asked Questions about beads

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

FAQPage Schema
How do I preserve AI agent memory across conversation sessions?

Beads provides persistent memory for AI agents by storing task state in a git-backed .beads directory, enabling context to survive session resets and context compaction. Use bd init to create memory, then bd ready to fetch tasks across sessions.

Can I track task dependencies and blockers across long-running projects?

Yes. Beads builds a graph-based dependency model that tracks task readiness, blockers, and relationships. Use bd dep to view dependency graphs and bd ready to filter tasks by completion status and prerequisites.

What's the best way to decompose epics into tracked subtasks with audit trails?

Beads integrates with Conductor to map epics to versioned tasks, then logs all decisions, notes, and status changes in audit trails. Use bd update to record changes and bd show to review the full history.

Do I need a database to store Beads state, or can I use version control?

Beads is git-backed by default, storing state as versioned .jsonl files in .beads. This eliminates database setup while preserving full audit history and enabling collaborative workflows.

How does Beads handle feature roadmaps that span multiple AI agent sessions?

Beads survives context compaction by storing prioritized tasks, labels, and dependency graphs persistently. Agents resume with full roadmap context via bd ready, maintaining continuity across disconnected sessions.