continue-tasks

Orchestrate queued project tasks by spawning agents and monitoring completion sentinels.

1|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/aberrantCode/llm_skills --skill continue-tasks
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: continue-tasks
Source: https://github.com/aberrantCode/llm_skills/tree/main/claude/skills/project-manager/sub-skills/continue-tasks
Command: npx skills add https://github.com/aberrantCode/llm_skills --skill continue-tasks

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the end-to-end orchestration of project tasks in multi-agent HomeRadar workflows to eliminate manual task selection, agent dispatch, and completion polling.

Core Features & Use Cases

  • Full Orchestration Loop: Performs bootstrap checks, generates missing plans, finds the next todo, writes the task file, spawns the appropriate agent, and repeats until all tasks are complete.
  • Agent Dispatch & Monitoring: Selects and launches the correct agent for each task and watches for task completion sentinels to drive the loop.
  • Use Case: In a HomeRadar project with queued todos and agent-based workers, use this Skill to continuously progress the project without manual intervention.

Quick Start

Use the project-manager skill to continue the workflow by invoking the /continue-tasks command for the current HomeRadar project.

Frequently Asked Questions about continue-tasks

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

FAQPage Schema
How do I automate task orchestration in a multi-agent project workflow?

Task orchestration in a multi-agent workflow is automated by sequentially picking the next todo, spawning the appropriate agent, and polling for completion sentinels until all queued tasks finish. This eliminates manual dispatch and monitoring.

What is the loop process for dispatching agents to queued todo tasks?

The loop process for dispatching agents involves bootstrap checks, generating missing plans, finding the next todo, writing the task file, and spawning the designated agent. It repeats this cycle continuously until every queued task reaches completion.

Do I need a project-manager context to run task automation loops?

Yes, task automation loops require access to the project-manager context, task plans, and task files. The environment must also support creating and monitoring agent processes or sentinel artifacts for proper lifecycle management.

How do I monitor agent lifecycle and task completion in project workflows?

Agent lifecycle and task completion are monitored by watching for task completion sentinels. These sentinel artifacts signal when a spawned agent finishes its assigned task, allowing the orchestration loop to automatically proceed to the next queued item.

Can I use workflow monitoring to continuously progress HomeRadar projects without manual intervention?

Yes, workflow monitoring continuously progresses HomeRadar projects without manual intervention by automatically iterating through queued todos. It selects the appropriate agent for each task and loops until the entire project task plan is complete.