AgentLayer▸docs
Reference

Naming your agent

The plugin is agent-kevin. Your agent does not have to be called Kevin.

Two separate things:

Lives inChangeable
Display name: what it calls itself, what the dashboard showsIDENTITY.md → - **Name:**Freely, any time
Namespace: the agent-kevin command prefix on every host, MCP tool namesThe plugin manifestsOnly by forking

init asks for a name, an emoji, and an avatar up front. Everything downstream reads the answer: templates are written in that name, the session banner and TASKS.md header render it, skill output like the standup card and the briefings signs off with its emoji, and the compile prompts refer to it, so your compiled memory speaks about the agent you actually named.

Renaming an existing home

rename-agent does the migration. It runs only when you type the command and asks for permission even then, because it rewrites files across the whole brain.

ChangesStays
IDENTITY.md: the Name (the old one added to Formerly), the emoji, the avatarThe home folder, and every path that contains it
Markdown the agent reads now: SOUL.md, AGENTS.md, USER.md, .claude/CLAUDE.md, rules, knowledge, projectsDated records: reports/, knowledge/raw/, every archive/ folder
TASKS.md and the dashboard, regeneratedCode, HTML and anything else that isn't markdown; your custom skills, which it lists for you to edit

It shows a dry run first: every line it would change, with the words around each mention, and the mentions it leaves alone, such as a path like Documents-Agents-Kevin.git. Read the lines, not the count: a mention that means someone else ("inspired by the minion Kevin") is in the list, and you keep it before anything is written.

The home folder keeps its name. Session history, the brain's version history and the home's settings are all tied to that path, so moving it is a separate job you do by hand.

Dated records keep the name they were written under. The - **Formerly:** line in IDENTITY.md loads every session and is read when knowledge compiles, so the agent knows the old name in them means itself. A seed bundle drops that line, since the agent it seeds has no such history. To finish a half-done rename, run it again naming the old name: that is a fresh dry run, so pass your keeps again.

Renaming survives updates: the upgrade skill resolves the name from your IDENTITY.md before diffing templates.

Forking is a different decision

Change name in the plugin manifests (.claude-plugin/plugin.json, .codex-plugin/plugin.json) and the env prefix follows (agent-scout reads SCOUT_*), giving you your own command namespace. The price is merging from upstream on every release, forever. If you only want the agent to feel like yours, rename it and skip the fork.

On this page