meto-cli 0.1.4 → 0.1.5

package/README.md

@@ -27,6 +27,7 @@ After running `meto-cli init`, your new project comes pre-loaded with everything
 - **Agent definitions** -- PM, developer, and tester agents configured to follow your methodology from day one
 - **Product context** -- your vision, tech stack, and decisions captured in structured files so every session starts with shared context
 - **Epics and workflows** -- definition of done, commit conventions, and an epic backlog to plan against
+- **Agent Teams ready** -- three agents configured to work in parallel with file ownership boundaries
 - **Token optimized** -- generated files are optimized for token consumption with Claude Code, so every session gets more productive output per dollar spent
 
 No more hours of manual setup. No more "I forgot to write the CLAUDE.md." It is all there from the start.
@@ -42,10 +43,11 @@ your-project/
 │   │   ├── developer-agent.md
 │   │   ├── pm-agent.md
 │   │   └── tester-agent.md
-│   └── agent-memory/
-│       ├── meto-developer/MEMORY.md
-│       ├── meto-pm/MEMORY.md
-│       └── meto-tester/MEMORY.md
+│   ├── agent-memory/
+│   │   ├── meto-developer/MEMORY.md
+│   │   ├── meto-pm/MEMORY.md
+│   │   └── meto-tester/MEMORY.md
+│   └── settings.json
 ├── ai/
 │   ├── backlog/
 │   │   └── epics.md
@@ -70,12 +72,33 @@ your-project/
 
 ---
 
+## Agent Teams
+
+Agent Teams is a Claude Code feature where multiple AI agents work in parallel on the same codebase, each with a specialized role.
+
+Meto scaffolds projects ready for Agent Teams out of the box:
+
+- **Three pre-configured agents** -- PM for planning, developer for building, tester for validation
+- **File ownership boundaries** -- each agent writes only to its designated files, preventing conflicts when running in parallel
+- **Feature enabled automatically** -- `.claude/settings.json` sets the experimental flag so Agent Teams works immediately
+
+**To activate:** start `claude` in your project, then say:
+
+> "Create an agent team with @meto-pm for planning, @meto-developer for building, @meto-tester for validation"
+
+**Display modes:** use Shift+Down to cycle between agents in-process, or run each agent in its own split pane (tmux/iTerm2).
+
+This feature is experimental and enabled via `.claude/settings.json` in the scaffold.
+
+---
+
 ## Next Steps
 
 1. `cd your-project`
 2. Open the project in your editor
 3. Start a Claude Code session and call `@meto-pm` to populate your backlog
-4. Pick your first task and start building
+4. Or spawn an agent team: tell Claude to create a team with @meto-pm, @meto-developer, and @meto-tester
+5. Pick your first task and start building
 
 ---
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "meto-cli",
-  "version": "0.1.4",
+  "version": "0.1.5",
   "description": "Scaffold structured software projects with built-in methodology",
   "type": "module",
   "bin": {

package/templates/.claude/agents/developer-agent.md

@@ -29,6 +29,12 @@ Update `.claude/agent-memory/meto-developer/MEMORY.md` with anything worth remem
 - Add features not in the acceptance criteria
 - Commit with `console.log`, `any` types, or commented-out code
 
+## Parallel Operation
+When running as a teammate: you read CLAUDE.md and this file fresh -- you do NOT have the lead's conversation history.
+Only write files listed under "What I Own".
+When a task is ready for testing, message @meto-tester directly (e.g., "tell @meto-tester slice-X is in testing").
+Never write to `/ai/backlog/`, `/ai/context/`, `tasks-backlog.md`, `tasks-todo.md`, or `tasks-done.md`.
+
 ## Task Pickup Protocol
 1. Read `tasks-todo.md` — take TOP item
 2. Copy full task block to `tasks-in-progress.md`, add `Started: [date]`

package/templates/.claude/agents/pm-agent.md

@@ -38,6 +38,12 @@ If `/ai/backlog/epics.md` has no real epics yet:
 - Make technical architecture decisions
 - Run bash commands
 
+## Parallel Operation
+When running as a teammate: you read CLAUDE.md and this file fresh -- you do NOT have the lead's conversation history.
+Only write files listed under "What I Own".
+If you need developer or tester action, message them directly (e.g., "tell @meto-developer the backlog is ready").
+Never write to `/src/`, `tasks-in-progress.md`, `tasks-in-testing.md`, or `tasks-done.md`.
+
 ## Task Definition Format
 
 ```markdown

package/templates/.claude/agents/tester-agent.md

@@ -28,6 +28,13 @@ Update `.claude/agent-memory/meto-tester/MEMORY.md` with patterns worth remember
 - Process items in parallel — always sequential
 - Skip any validation check
 
+## Parallel Operation
+When running as a teammate: you read CLAUDE.md and this file fresh -- you do NOT have the lead's conversation history.
+Only write files listed under "What I Own".
+When validation is complete, message the lead or @meto-developer directly with the result.
+Never write to `/src/`, `/ai/backlog/`, `/ai/context/`, `tasks-backlog.md`, or `tasks-in-progress.md`.
+Process items sequentially even when other agents run in parallel.
+
 ## Validation Protocol
 ONE item at a time — parallel writes corrupt the board. Always sequential.
 

package/templates/.claude/settings.json

@@ -0,0 +1,5 @@
+{
+  "env": {
+    "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1"
+  }
+}

package/templates/CLAUDE.md

@@ -44,6 +44,26 @@ See `/ai/workflows/definition-of-done.md` for done criteria.
 
 ---
 
+## Agent Teams
+
+This project supports Agent Teams. The lead agent can spawn teammates using `@meto-pm`, `@meto-developer`, `@meto-tester`.
+
+**Coordination model:** Agent Teams has its own task system, but this project uses the kanban board (`tasks-backlog` through `tasks-done`) as the single source of truth for task state. Teammates must read and update the board files, not rely on Agent Teams' internal task tracking.
+
+**File ownership is exclusive -- two teammates editing the same file causes overwrites.**
+
+| Agent | Writes |
+|---|---|
+| `@meto-pm` | `/ai/` files, `tasks-backlog.md`, `tasks-todo.md` |
+| `@meto-developer` | `/src/`, config files, `tasks-in-progress.md`, `tasks-in-testing.md` |
+| `@meto-tester` | `tasks-in-testing.md`, `tasks-done.md`, `tasks-todo.md` (failed items) |
+
+Each agent writes only its own memory file in `.claude/agent-memory/` -- never another agent's.
+
+Teammates do NOT inherit the lead's conversation history. Each teammate reads CLAUDE.md and its agent definition fresh.
+
+---
+
 ## Commit Format
 
 ```