Onboarding

What the init wizard asks, what it writes, and the one convention every session depends on.

The wizard

Run the init skill from a fresh session in the home (/agent-kevin:init in Claude Code; under Codex, create the home from Claude Code for now). It first checks the prerequisites and the plugin's packages, and stops with the exact fix if one is missing.

❓ What should this agent be called? (default: Kevin; plus emoji + avatar if you rename)
❓ Starting from a teammate's seed bundle? (applied after the scaffold)
❓ Its character (SOUL): accept default, or refine?
❓ Its role (IDENTITY): general / coding / research / planning / custom?
❓ Your name and home timezone?
❓ Where does your code live? (optional)
❓ Paste any URLs about you (blog, GitHub, talks) so Kevin seeds your profile
❓ Paste a path or URL for your avatar (optional)
❓ Should knowledge/ and projects/ live somewhere else?
❓ Your world: current projects, this year's goals, the company this agent works for
❓ Drop documents into the inbox: your LinkedIn PDF, résumé, company one-pagers
❓ Communication style and any hard preferences?
❓ Signal topics the briefings should watch?
❓ Which model: Opus or Fable?
❓ Keep a history of the agent's memory, so changes can be undone?
❓ Do you also run Codex from this home?
❓ Configure skill packs (SEO, Browser, third-party libraries)?
❓ Confirm + scaffold

About twenty minutes, most of it the part about you and your world. Every question past the name can be skipped. Each answer becomes the default for later steps, the name first of all, since every file the wizard writes is phrased in it. See Naming your agent. A seed bundle from a teammate or the web wizard answers most of it for you.

Your world

The questions the web wizard asks, answered in the session instead:

You give	Init writes
The projects you're working on	`projects/<slug>/README.md` and a `tasks/` folder for each
This year's goals	`knowledge/concepts/roadmap-draft.md`, which the `roadmap` skill turns into `roadmap.html`
The company, its website, your role	`knowledge/concepts/company-profile.md`, with the website read in
Documents a new hire would read	Files in `knowledge/raw/inbox/`: init reads what's about you and the company straight away, and the first `knowledge-compile` absorbs the rest

LinkedIn blocks automated reads, so a profile link gives the agent nothing. Open your profile, choose More → Save to PDF, and drop the file in the inbox when init asks. It's the fastest way to hand over your career history.

What it writes

CLAUDE.md

settings.json

settings.local.json

AGENTS.md

SOUL.md

IDENTITY.md

USER.md

File	Role
`AGENTS.md`	The operating manual. Harness-neutral: every AGENTS.md-aware CLI reads it
`SOUL.md` / `IDENTITY.md` / `USER.md`	Kevin's character, Kevin's role, your headline
`knowledge/`, `projects/`	The brain and the work, optionally at custom locations
`assets/`	Kevin's avatar and any other media for the whole home. Your own picture goes in `knowledge/user/assets/`
`scripts/`	Scripts you keep for recurring tasks, with your browser flows in `scripts/browser/`, one folder each
`.claude/CLAUDE.md`	The Claude Code bridge: `@`-imports the manual and the identity stack
`.claude/settings.json`	The home's permissions: marketplace registration plus grants for the always-on core tools. Read by Claude Code directly and by Codex through its generated wiring
`.claude/settings.local.json`	Gitignored. Non-secret config only; secrets go in `.state/secrets/.env`
`.claude/rules/`	Path-scoped coding rules (TypeScript, Swift), applied when a matching file is in play. Claude Code reads them; Codex doesn't
`.codex/`	Only when you said you run Codex: hooks, MCP registration, permission profile, rules. See Codex

You fill secret values in your editor, never in chat. Neither init nor configure-skills asks for a key in conversation, because pasted secrets touch the transcript.

If you moved knowledge/ or projects/ outside the home, the wizard appends the needed permission entries so the host can reach them without prompting.

Exit and relaunch

Plugin registration only takes effect on a fresh session.

/exit
cd ~/Documents/Agents/Kevin && claude

Watch for a marketplace trust prompt on first relaunch and accept it. If you miss it, recover with /plugin install agent-kevin --marketplace AgentLayer1/agentlayer-agent-marketplace.

The one convention

Always launch from the agent home, then reach your code from inside the session.

The plugin is enabled by the home's settings, so a host loads it only for sessions started there. That single fact gives you the identity stack, the skills, the MCP tools, and session capture. It is also what keeps several agents on one machine apart with no configuration.

Reaching your code. Add the repos you work in to permissions.additionalDirectories in the home's settings.local.json. Both hosts honour it: Claude Code grants file access there, and Codex lists the same paths as workspace roots.

{
  "permissions": {
    "additionalDirectories": ["/Users/you/Developer/acme", "/Users/you/Developer/side-project"]
  }
}

Set AGENT_CODE_PATH to your primary repo and its git activity shows up in every session's context.

The session can wander, the anchor cannot. Once running, cd anywhere. The home is resolved from the launch directory, which never moves, so capture lands in the right brain even when the shell has roamed three repos deep.

What you give up

A session started outside a home is not captured, and says nothing about it. No plugin loads, so no hooks run. Do not put AGENT_HOME in a user-level settings file to work around this: it is machine-wide and hands every session to one brain. Per-home config belongs in that home's settings.local.json.