How do I install the Agentpack Operator skill?

Run npx skillmds@latest add tomevault-io/agentpack-operator in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Agentpack Operator safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Agentpack Operator?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Agentpack Operator free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Agentpack Operator?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Agentpack Operator by tomevault-io — Agent Skill

Name: Agentpack Operator
Author: tomevault-io

File contents

agentpack-operator

Operate Agentpack safely and reproducibly.

What you can do

Self-check environment: agentpack doctor --json
Update sources: agentpack update --yes --json
Preview changes: agentpack preview --json (or agentpack plan --json + agentpack diff --json)
Apply changes: agentpack deploy --apply --yes --json
Verify drift: agentpack status --json
Explain why things look the way they do: agentpack explain plan|diff|status --json
Record outcomes and score modules: agentpack record + agentpack score
Propose overlay updates from drift: agentpack evolve propose --yes --json
Roll back: agentpack rollback --to <snapshot_id>

Safety rules

Never run deploy --apply unless the user explicitly asked to apply.
Always show the plan + (optional) diff summary first.
Prefer --json for machine parsing; keep outputs small and scoped.
Use --repo, --profile, --target when operating outside defaults.

Typical workflow

0) Doctor

agentpack doctor --json

0.5) Update (optional)

agentpack update --yes --json

1) Plan

agentpack plan --json

2) Diff (optional)

agentpack diff --json

3) Apply (only with user approval)

agentpack deploy --apply --yes --json

4) Status

agentpack status --json

AI-first feedback loop (optional)

Explain (optional)

agentpack explain status --json

Record + score (optional)

Record a lightweight execution event (example):

echo '{"module_id":"command:ap-deploy","success":true}' | agentpack record
agentpack score

Evolve overlays (optional)

Capture drifted deployed files into overlays (creates a local git branch):

agentpack evolve propose --yes --json

Converted and distributed by TomeVault — claim your Tome and manage your conversions.

---name: agentpack-operatordescription: Operate agentpack safely (doctor/update/preview/diff/deploy/status/explain/evolve/rollback). Prefer --json. Use when this capability is needed.---# agentpack-operatorOperate Agentpack safely and reproducibly.## What you can do- Self-check environment: `agentpack doctor --json`- Update sources: `agentpack update --yes --json`- Preview changes: `agentpack preview --json` (or `agentpack plan --json` + `agentpack diff --json`)- Apply changes: `agentpack deploy --apply --yes --json`- Verify drift: `agentpack status --json`- Explain why things look the way they do: `agentpack explain plan|diff|status --json`- Record outcomes and score modules: `agentpack record` + `agentpack score`- Propose overlay updates from drift: `agentpack evolve propose --yes --json`- Roll back: `agentpack rollback --to <snapshot_id>`## Safety rules1. Never run `deploy --apply` unless the user explicitly asked to apply.2. Always show the `plan` + (optional) `diff` summary first.3. Prefer `--json` for machine parsing; keep outputs small and scoped.4. Use `--repo`, `--profile`, `--target` when operating outside defaults.## Typical workflow### 0) Doctor```bashagentpack doctor --json```### 0.5) Update (optional)```bashagentpack update --yes --json```### 1) Plan```bashagentpack plan --json```### 2) Diff (optional)```bashagentpack diff --json```### 3) Apply (only with user approval)```bashagentpack deploy --apply --yes --json```### 4) Status```bashagentpack status --json```## AI-first feedback loop (optional)### Explain (optional)```bashagentpack explain status --json```### Record + score (optional)Record a lightweight execution event (example):```bashecho '{"module_id":"command:ap-deploy","success":true}' | agentpack recordagentpack score```### Evolve overlays (optional)Capture drifted deployed files into overlays (creates a local git branch):```bashagentpack evolve propose --yes --json```---> Converted and distributed by [TomeVault](https://tomevault.io/claim/liqiongyu) — claim your Tome and manage your conversions.

Frequently asked questions

Agentpack Operator