start-pairing

A pair-programming session manager that records who is driving, who is navigating, decisions, and lessons learned. Pair programming is when two developers work together on the same task.

In plain words
What is it for?
Use it to start, end, or list pairing sessions, link one to a Jira ticket or topic, share a summary in Slack, or leave notes for remote pairing.
Why use it?
It gives the session a shared record, so knowledge and decisions are easier to pass between partners and review later.

Command for Claude Code

Part of the claude-workflow plugin — 10 skills, 25 commands, 3 agents, 6 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/parkerm2/create-claude-workflow/start-pairing
Clone the repo
git clone --depth 1 https://github.com/ParkerM2/create-claude-workflow

Made for: Claude Code.

Or install claude-workflow, the plugin that ships this one along with the rest of its 10 skills, 25 commands, 3 agents, 6 hooks.

Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,648 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00013 $0.04648
Opus 5 $0.00006 $0.02324
Sonnet 5 $0.00003 $0.00930
Haiku 4.5 $0.00001 $0.00465

Measured 3d ago against content hash 3e18a0cbb0d8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

start-pairing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/start-pairing.md · 685 lines

How it starts

The opening of the file, as written. The whole thing — 685 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Start Pairing Command

Manages pair programming sessions with structured knowledge transfer. Tracks driver/navigator roles, captures decisions and learnings, and produces session summaries.

Usage

/start-pairing [TICKET_ID] [--topic TOPIC] [--end] [--slack] [--remote]
/start-pairing --end                    # End active pairing session
/start-pairing --list                   # List recent pair sessions

Options:

  • TICKET_ID: Jira ticket ID for context (optional, topic required if omitted)
  • --topic TOPIC: Free-form topic name if not ticket-based
  • --end: End the current active session
  • --slack: Post session summary to Slack
  • --remote: Async pairing mode (leave notes for partner)
  • --list: Show recent pairing sessions

Phase 1: Session Setup

// Check for active session
const getActiveSession = () => {
  const sessionDir = '.claude/sessions';
  if (!fs.existsSync(sessionDir)) {
    return null;
  }

  const files = fs.readdirSync(sessionDir)
    .filter(f => f.startsWith('pair-') && f.endsWith('.md'));

  if (!files.length) return null;

  // Get most recent
  const latest = files.sort().pop();
  const content = fs.readFileSync(`${sessionDir}/${latest}`, 'utf8');

  // Check if session is still active (no end marker)
  if (!content.includes('## Session End')) {
    return {
      file: `${sessionDir}/${latest}`,
      data: content
    };
  }

  return null;
};

// Handle --end flag
if (args.includes('--end')) {
  const active = getActiveSession();
  if (!active) {
    console.log('❌ No active pairing session found');
    return;
  }

  console.log('🛑 Ending pairing session...');
  // Jump to Phase 4
  await endSession(active.file);
  return;
}

// Handle --list flag
if (args.includes('--list')) {
  const sessionDir = '.claude/sessions';
  if (!fs.existsSync(sessionDir)) {
    console.log('No session history');
    return;
  }

  const files = fs.readdirSync(sessionDir)
    .filter(f => f.startsWith('pair-') && f.endsWith('.md'))
    .sort()
    .reverse();

  console.log('\n📋 Recent Pairing Sessions:\n');
  files.slice(0, 10).forEach(f => {
    const content = fs.readFileSync(`${sessionDir}/${f}`, 'utf8');
    const match = content.match(/^# Pair Session: (.+)$/m);
    const title = match ? match[1] : f;
    console.log(`- ${f.replace('.md', '')}\n  ${title}`);
  });
  return;
}

// Detect active session
const activeSession = getActiveSession();
if (activeSession && !args[0]) {
  console.log('⚠️  Active pairing session detected:');
  console.log(activeSession.data.split('\n').slice(0, 5).join('\n'));
  console.log('Use /start-pairing --end to end it, or /start-pairing TICKET to start new');
  return;
}

console.log('\n🚀 Starting pair programming session...\n');

// Get ticket ID or topic
let ticketId = args[0];
let topic = null;

if (ticketId && !ticketId.match(/[A-Z]+-\d+/)) {
  topic = ticketId;
  ticketId = null;
} else if (args.includes('--topic')) {
  const idx = args.indexOf('--topic');
  topic = args[idx + 1];
}

if (!ticketId && !topic) {
  ticketId = await prompt('Ticket ID (or press enter for free topic): ');
  if (!ticketId) {
    topic = await prompt('Topic for this pairing session: ');
  }
}

// Get participants
console.log('\n👥 Pairing Participants\n');

const driver = await prompt('Driver name (who\'s coding): ');
if (!driver) {
  console.log('❌ Driver name required');
  return;
}

const navigator = await prompt('Navigator name (who\'s reviewing/guiding): ');
if (!navigator) {
  console.log('❌ Navigator name required');
  return;
}

// Get session goal
const goal = await prompt('\n🎯 Session goal or focus area: ');
const timeBox = await prompt('Time box in minutes (optional): ') || '60';

console.log('✓ Session configured\n');

// Create session directory
const sessionDir = '.claude/sessions';
if (!fs.existsSync(sessionDir)) {
  fs.mkdirSync(sessionDir, { recursive: true });
}

// Create session file
const timestamp = new Date().toISOString().replace(/[:.]/g, '-').slice(0, -5);
const sessionFile = `${sessionDir}/pair-${timestamp}.md`;

const sessionTitle = ticketId ? `${ticketId} - ${goal}` : `${topic} - ${goal}`;

const initialContent = `# Pair Session: ${sessionTitle}

**Started**: ${new Date().toISOString()}
**Driver**: ${driver}
**Navigator**: ${navigator}
**Time Box**: ${timeBox} minutes
${ticketId ? `**Ticket**: ${ticketId}` : `**Topic**: ${topic}`}

## Session Context

(Loading context...)

## Decisions Made

(Decisions will be tracked here)

## Code Changes

(Summary of code modifications)

## Learnings

(Key learnings and patterns discovered)

## Open Questions

(Unresolved issues or follow-ups)

## Session Notes

(Ongoing session notes)

`;

fs.writeFileSync(sessionFile, initialContent);
console.log(`📝 Session file: ${sessionFile}\n`);

// Store active session reference
const activeFile = `${sessionDir}/.active`;
fs.writeFileSync(activeFile, sessionFile);

Read the full file on GitHub · 685 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 685 lines · 13 tokens per session scan A 3e18a0cbb0d8

Subscribe to this mod's changes

start-pairing is a command published in the GitHub repository ParkerM2/create-claude-workflow (4 stars, last pushed 5mo ago), licensed MIT. It adds 13 tokens to every session and 4,648 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.