Skip to content
Development
Command

/system-upgrade

Skip Phase 3 approval gate and apply all changes automatically

From plugin
vexjoy-agent
42111 skills198 agents11 commands76 hooks
Install
$ npx -y skills add notque/vexjoy-agent --agent claude-code

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/system-upgrade

Context preview

What this command does when you run it.

Skip Phase 3 approval gate and apply all changes automatically

Command definition

system-upgrade.md
name: system-upgrade
description: |
  Systematic upgrade pipeline for adapting agents, skills, and hooks when
  Claude Code ships updates, user goals change, or retro learnings accumulate.
version: 1.0.0
route_to:
  agent: system-upgrade-engineer
  skill: system-upgrade
  enhancements:
    - agent-evaluation
    - routing-table-updater
trigger:
  keywords:
    - upgrade agents
    - system upgrade
    - claude update
    - upgrade skills
    - adapt workflow
    - apply claude update
    - apply retro
parameters:
  required:
    - name: trigger
      description: |
        What triggered the upgrade: Claude Code version/release notes,
        description of goal change, or "retro" to use learning.db graduation candidates
  optional:
    - name: scope
      description: "comprehensive | default (default: recent 10 agents + all hooks + all routing tables)"
      default: "default"
    - name: auto
      description: "Skip Phase 3 approval gate and apply all changes automatically"
      default: "false"

/system-upgrade

Entry point for the System Upgrade Pipeline.

Usage

/do upgrade the system — claude just shipped [feature]
/do system upgrade — I'm now working with [new domain]
/do apply retro learnings to upgrade the system

What It Does

**6-Phase Pipeline:**

1. **CHANGELOG** — Parse the trigger, extract actionable change signals 2. **AUDIT** — Scan agents/skills/hooks for affected components 3. **PLAN** — Rank changes (Critical/Important/Minor), present to user for approval 4. **IMPLEMENT** — Dispatch domain specialists in parallel for approved changes 5. **VALIDATE** — Score modified components before/after with agent-evaluation 6. **DEPLOY** — Commit, sync to `~/.claude`, create PR

Three Trigger Types

| Type | When | Input | |------|------|-------| | **claude-release** | Claude Code ships new version | "Claude added Notification hook event" | | **goal-change** | Your workflow/focus shifts | "I'm now working with Rust" | | **retro-driven** | Retro learnings ready to embed | "Apply retro" or "/retro graduate" output |

Key Behavior

**Phase 3 always gates on your approval.** The pipeline shows you a ranked table of proposed changes with effort estimates before touching any files. You can approve all, pick specific items, or stop.

Examples

/do upgrade the system — Claude Code shipped structured output support for tools
/do system upgrade — I want to add Rust support to the agent system
/do apply retro learnings to upgrade the system
/do claude update — new hook event Notification was added
Read more
Ships withvexjoy-agent

Essays and writing behind this toolkit live at vexjoy.com. VexJoy Agent connects plain-English requests to specialist agents, skills, and workflows. /do selects the knowledge and tools needed for your task.

Get the whole plugin

Other commands on vexjoy-agent.