agentbase

Orchestrates autonomous agents to coordinate multi-workstream development projects.

1|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/sidmohan0/agentbase --skill agentbase
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agentbase
Source: https://github.com/sidmohan0/agentbase/tree/main/skill
Command: npx skills add https://github.com/sidmohan0/agentbase --skill agentbase

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires git, claude, tmux, npm, and includes scripts (resource) components.

What problem does it solve?

This Skill provides a structured, automated orchestration layer to manage multiple autonomous agents coordinating work across codebases. It reduces manual coordination and accelerates multi-workstream development by organizing planners, workers, and a judge into a cohesive workflow.

Core Features & Use Cases

  • Agent Planner orchestration: Spawns and manages autonomous planner agents to discover, triage, and assign tasks across workstreams.
  • Worker orchestration: Creates worker sub-agents to execute tasks in parallel while respecting workstream ownership.
  • Progress monitoring & governance: Supports graceful stop/resume and periodic reporting to a human master orchestrator.
  • Use Case: Manage frontend, backend, and API workstreams in parallel to deliver features faster with clear ownership.

Quick Start

  1. Install the skill into Claude Code following standard Quick Start steps.
  2. Run /agentbase init to scaffold or adapt your repo, then /agentbase go to start planning.
  3. Use /agentbase status and /agentbase stop to monitor and control execution.

Frequently Asked Questions about agentbase

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

FAQPage Schema
How do I orchestrate multiple autonomous agents to work on different software workstreams in parallel?

Multi-agent orchestration coordinates autonomous planner and worker sub-agents to execute parallel development workstreams. This Skill manages task triage, assignment, and lifecycle tracking across codebases to accelerate structured feature delivery.

What do I need to set up before spawning autonomous sub-agents for task management?

Autonomous sub-agent task management requires a Unix-like shell environment with Claude Code CLI, git, tmux, and npm installed. These dependencies allow the orchestration scripts to spawn agents, manage their execution environments, and track workstream progress.

How does multi-agent orchestration handle task discovery and execution across frontend and backend workstreams?

Multi-agent orchestration uses a planner agent to discover and triage tasks, then spawns worker sub-agents to execute them in parallel. Workers respect workstream ownership boundaries while a judge agent monitors progress and reports back to the human orchestrator.

Can I pause and resume multi-agent workflow execution without losing track of task progress?

Multi-agent workflow execution supports graceful stop and resume functionality. You can run status checks to monitor governance and periodically report progress to the human master orchestrator, ensuring controlled execution across all active workstreams.

Is there a way to automate project triage and planning before starting parallel development workstreams?

Automated project triage and planning begins by initializing the orchestration framework to scaffold your repository, then triggering the planner agents to discover, assign, and sequence tasks across multiple workstreams before spawning parallel worker execution.

Why use a dedicated orchestration layer instead of manually coordinating multiple Claude Code agents?

A dedicated orchestration layer reduces manual coordination overhead by structuring the lifecycle of multiple agents. It enforces workstream ownership, parallel execution, and automated reporting, which prevents task conflicts and accelerates multi-workstream delivery.