agent

Load agent lifecycle skills to orchestrate tasks in local or dispatched modes.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/matt-metivier/zk-hub --skill agent-matt-metivier
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent
Source: https://github.com/matt-metivier/zk-hub/tree/main/skills/agent
Command: npx skills add https://github.com/matt-metivier/zk-hub --skill agent-matt-metivier

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Agent lifecycle skills enable a CC-based agent to conduct end-to-end zk-hub task lifecycles without a dedicated agent daemon, coordinating research, synthesis, design, evidence, briefs, and patterns.

Core Features & Use Cases

  • Orchestrate the full task lifecycle in two modes: local (cc-driven) and dispatched (hub-driven).
  • Load persona, repo conventions, and domain skills to guide execution.
  • Persist a trace of decisions, artifacts, and evidence for audits and vaulting.

Quick Start

Activate the agent lifecycle skill and begin a local or dispatched task flow with a single instruction.

Frequently Asked Questions about agent

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

FAQPage Schema
How do I orchestrate task lifecycles without a dedicated agent daemon?

Task lifecycle orchestration without a dedicated agent daemon is achieved by loading agent lifecycle skills to coordinate phases directly. This enables execution in local or dispatched modes, managing research, design, and evidence without persistent background processes.

What is the difference between local and dispatched workflow modes?

Local workflow mode is cc-driven for direct execution, while dispatched mode is hub-driven for remote task management. Both modes load persona, repo conventions, and domain skills to guide the full task lifecycle without requiring a dedicated agent process.

How do I load domain skills and repo conventions for workflow orchestration?

Loading domain skills and repo conventions requires identifying and loading agent lifecycle skills from the repository. This automatically imports machine persona, specific repository conventions, domain skills, and patterns to guide execution throughout the task lifecycle.

Can I persist decision traces and evidence for audits during task orchestration?

Decision traces and evidence for audits are persisted automatically during task lifecycle orchestration. The workflow captures decisions, artifacts, and evidence throughout execution, providing a complete audit trail for compliance and vaulting purposes.

Do I need a dedicated agent process to manage research, synthesis, and design phases?

A dedicated agent process is not needed to manage research, synthesis, and design phases. The agent lifecycle skill enables a cc-based agent to conduct end-to-end task coordination by loading the required skills and operating in local or dispatched modes.

What limitations exist when coordinating end-to-end task lifecycles without an agent?

Coordinating end-to-end task lifecycles without an agent requires loading skills from the repository first. Limitations include dependency on proper repo convention loading and the need to operate within either local or dispatched modes for lifecycle phase management.