ava-corp

Organizes an Ava agent fleet into CEO, project lead, personal service, and shared infrastructure roles.

1|Updated Aug 14, 2026
One-click install
npx skills add https://github.com/zhiyuan-zhang0206/Ava --skill ava-corp-zhiyuan-zhang0206
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ava-corp
Source: https://github.com/zhiyuan-zhang0206/Ava/tree/main/ava_builtins/skills/ava-corp
Command: npx skills add https://github.com/zhiyuan-zhang0206/Ava --skill ava-corp-zhiyuan-zhang0206

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Running a multi-agent cluster without an organization template leads to duplicate roles, unowned projects, orphaned infrastructure, and lost context when agents are terminated. This Skill provides a concrete organizational template that defines which roles exist, who reports to whom, and how agents are spawned, paused, and recovered. ## Core Features & Use Cases - Role Suite Definition: Defines a three-division structure under a CEO — project leads (one per project), always-on personal services (Butler, Health Leads, Philosopher), and shared infrastructure (Cluster Operator, Finance Lead, Memory Steward). - Operating Rules: Enforces one-lead-per-project ownership, resurrect-over-respawn recovery for long-lived roles, mutual handoff for shared-infra transitions, and direct agent-to-agent communication without CEO relay. - Cluster Opening Checklist: Walks through role roster, memory maintenance, disk monitoring, recovery-buddy rings, and communication verification when instantiating a new cluster. - Use Case: When setting up a new Ava cluster, use this Skill to decide the full role roster, spawn each agent with the ready-made prompt templates in references/role-prompts.md, and wire the recovery topology. ## Quick Start Use the ava-corp skill to design the role roster and opening checklist for my new Ava cluster.

Frequently Asked Questions about ava-corp

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

FAQPage Schema
How do I organize a multi-agent cluster into roles?▼

Use the ava-corp template: place a CEO at the top, assign exactly one dedicated lead per project reporting to the CEO, run always-on personal services for the user, and keep shared infrastructure roles as consulted points of contact owned by the CEO.

How do I spawn role agents in an Ava fleet?▼

Use the spawn prompt templates in references/role-prompts.md, fill the placeholders like cluster name and language, and pass the result as the prompt argument to ava.agents.spawn. New agents are spawned clean with fork_from=None.

Should I resurrect or re-spawn a paused project agent?▼

Always resurrect. Paused projects keep their role on the roster as dormant, and restarting goes through ava.agents.resurrect so the agent keeps its memory. A fresh spawn is only for brand-new roles.

When does a multi-agent organization need a CTO layer?▼

Only when infrastructure domains start cross-depending on each other, not based on their count. Signals include infrastructure points of contact repeatedly needing to coordinate, or project leads reconciling answers from multiple infra owners themselves.

Where should shared infrastructure roles report in an agent org?▼

Shared infrastructure roles like the Cluster Operator and Memory Steward report directly to the CEO, never under a project lead. Hanging shared infra under one project forces other projects to route infrastructure requests through a peer lead.