executing-plans

## Collab Session Required

Before proceeding, check for active collab session:

1. Check if `.collab/` directory exists
2. Check if any session folders exist within
3. If no session found:
   ```
   No active collab session found.

   Use /collab to start a session first.
   ```
   **STOP** - do not proceed with this skill.

4. If multiple sessions exist, check `COLLAB_SESSION_PATH` env var or ask user which session.

## Browser-Based Questions

When a collab session is active, use `render_ui` for all user interactions.

**Component selection:**
| Question Type | Component |
|--------------|-----------|
| Yes/No | Card with action buttons |
| Choose 1 of 2-5 | RadioGroup |
| Choose 1 of 6+ | MultipleChoice |
| Free text | TextInput or TextArea |

**Example - Yes/No:**
```
Tool: mcp__plugin_mermaid-collab_mermaid__render_ui
Args: {
  "project": "<cwd>",
  "session": "<session>",
  "ui": {
    "type": "Card",
    "props": { "title": "<question context>" },
    "children": [{ "type": "Markdown", "props": { "content": "<question>" } }],
    "actions": [
      { "id": "yes", "label": "Yes", "primary": true },
      { "id": "no", "label": "No" }
    ]
  },
  "blocking": true
}
```

**Terminal prompts only when:** No collab session exists (pre-session selection).

# Executing Plans

## Overview

Load plan, review critically, execute tasks in batches, report for review between batches.

**Core principle:** Batch execution with checkpoints for architect review.

**Announce at start:** "I'm using the executing-plans skill to implement this plan."

## The Process

### Step 1: Load and Review Plan
1. Read plan file
2. Review critically - identify any questions or concerns about the plan
3. If concerns: Raise them with your human partner before starting
4. If no concerns: Create TodoWrite and proceed

### Step 1.1: Parse Task Dependency Graph (Collab Workflow)

When called from `rough-draft` within a collab workflow, the plan includes a task dependency graph. Parse and prepare