issue-start

Issue Start

Begin work on a GitHub issue with parallel agents based on work stream analysis.

Usage

/pm:issue-start <issue_number>

Quick Check

1. Get issue details:

   gh issue view $ARGUMENTS --json state,title,labels,body
   

   If it fails: "❌ Cannot access issue #$ARGUMENTS. Check number or run: gh auth login"

2. Find local task file:

   • First check if .claude/epics/*/$ARGUMENTS.md exists (new naming)
   • If not found, search for file containing github:.*issues/$ARGUMENTS in frontmatter (old naming)
   • If not found: "❌ No local task for issue #$ARGUMENTS. This issue may have been created outside the PM system."

3. Check for analysis:

   test -f .claude/epics/*/$ARGUMENTS-analysis.md || echo "❌ No analysis found for issue #$ARGUMENTS
   
   Run: /pm:issue-analyze $ARGUMENTS first
   Or: /pm:issue-start $ARGUMENTS --analyze to do both"
   

   If no analysis exists and no --analyze flag, stop execution.

Instructions

1. Ensure Worktree Exists

Check if epic worktree exists:

# Find epic name from task file
epic_name={extracted_from_path}

# Check worktree
if ! git worktree list | grep -q "epic-$epic_name"; then
  echo "❌ No worktree for epic. Run: /pm:epic-start $epic_name"
  exit 1
fi

2. Read Analysis

Read .claude/epics/{epic_name}/$ARGUMENTS-analysis.md:

• Parse parallel streams
• Identify which can start immediately
• Note dependencies between streams

3. Setup Progress Tracking

Get current datetime: date -u +"%Y-%m-%dT%H:%M:%SZ"

Create workspace structure:

mkdir -p .claude/epics/{epic_name}/updates/$ARGUMENTS

Update task file frontmatter updated field with current datetime.

4. Launch Parallel Agents

For each stream that can start immediately:

Create .claude/epics/{epic_name}/updates/$ARGUMENTS/stream-{X}.md:

---
issue: $ARGUMENTS
stream: {stream_name}
agent: {agent_type}
started: {current_datetime}
status: in_progress
---

# Stream {X}: {stream_name}

## Scope
{stream_description}

## Files
{file_patterns}

## Progress
- Starting implementation