beads

Track tasks and dependencies across sessions with the Beads issue tracker.

564|70|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/ynulihao/AgentSkillOS --skill beads-ynulihao
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: beads
Source: https://github.com/ynulihao/AgentSkillOS/tree/main/data/skill_seeds/beads
Command: npx skills add https://github.com/ynulihao/AgentSkillOS --skill beads-ynulihao

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bd, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a robust, persistent task tracking system that survives conversation compaction and manages complex dependencies, ensuring work context is never lost.

Core Features & Use Cases

  • Persistent Memory: Tracks tasks across multiple sessions and compaction cycles.
  • Dependency Management: Visualizes and enforces task dependencies, automatically surfacing ready work.
  • Session Handoff: Writes notes that allow agents to resume work with zero context loss.
  • Use Case: An AI agent working on a multi-week software development project can use this Skill to track features, bugs, and their interdependencies, ensuring progress is maintained even after long breaks or context resets.

Quick Start

Use the beads skill to create a new task for implementing user authentication.

Frequently Asked Questions about beads

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

FAQPage Schema
How do I maintain persistent task memory for AI agents across conversation compaction?

Persistent task memory for AI agents is maintained by using an issue tracker with dependency graphs that survives conversation compaction. This approach stores multi-session context, ensuring work progress is never lost during context resets.

How do I track complex task dependencies in multi-session AI workflows?

Track complex task dependencies in multi-session AI workflows by visualizing and enforcing dependency graphs. This system automatically surfaces ready work and writes session handoff notes, allowing agents to resume with zero context loss.

What is the best way to manage issue tracking when work spans multiple agent sessions?

Issue tracking for work spanning multiple agent sessions is best managed using persistent memory that writes handoff notes. This ensures the agent can resume work with zero context loss even after long breaks or compaction cycles.

Can I use bash commands to query and update task status in a dependency graph?

Yes, you can use bash commands to query and update task status in a dependency graph. The system allows Read and Bash tools for the bd dependency to create tasks, show tasks, and check what is blocking work.

How do I find what tasks are ready to work on or what is blocking my AI agent?

Find ready tasks or blocking issues by triggering the system with phrases like what's ready to work on or what's blocking. It evaluates the dependency graph to surface actionable tasks and identify blockers for the AI agent.