Quarterback

Read the field. Call the play.

Strategic task prioritization and agent orchestration for multi-project operators.

Every other AI task manager breaks down one project into subtasks. Quarterback helps you decide which of your ten projects to prioritize right now — using a 5-factor weighted scoring engine, organizational context, and time-aware planning. It runs locally, costs nothing, and works as both a standalone CLI and an MCP server for Claude.

What Makes Quarterback Different

Quick Start

# Install
pip install quarterback

# Initialize (creates ~/.quarterback/)
quarterback init

# Interactive setup wizard — walks you through org, goals, workflows, projects, constraints
quarterback setup

# Add your first project and tasks
quarterback add "Launch landing page" --project "My Startup" --priority 4 --effort 3 --impact 5
quarterback add "Write blog post" --project "Content" --priority 3 --effort 2 --impact 3

# See what to work on
quarterback priorities

# Find quick wins
quarterback quick-wins

# Plan your day with time awareness
quarterback plan-day

LLM-Powered Setup (via MCP)

When using Quarterback as an MCP server, ask your LLM: "Set up Quarterback for me" — it will call the setup_quarterback tool, interview you conversationally about your business, goals, workflows, projects, constraints, and knowledge base (Playbook), then write all config files and database records in one shot. No manual YAML editing required.

MCP Server

Quarterback works with any MCP-compatible client — Claude Desktop, Claude Code, Cursor, Windsurf, Cline, OpenAI agents, and others. All 23 tools use standard MCP protocol (JSON-RPC over stdio) with no LLM-specific dependencies.

# Install with MCP support
pip install quarterback[mcp]

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "quarterback": {
      "command": "quarterback-server"
    }
  }
}

Or for Claude Code (~/.claude/settings.json):

{
  "mcpServers": {
    "quarterback": {
      "command": "quarterback-server"
    }
  }
}

The same quarterback-server command works with any MCP client — just add it to your client's server configuration.

Then ask your LLM: "What should I work on today?" — it will use all 23 Quarterback tools to analyze your priorities.

Features

5-Factor Prioritization Engine

Every task is scored across five dimensions:

Advisory Document System

Analyze external articles, books, and advice against your organizational context:

# Import and auto-analyze an article
quarterback advisory-add --title "Growth Strategy" --url https://example.com/article

# Review the analysis
quarterback advisory-view --id 1

# Approve recommendations (optionally create tasks)
quarterback advisory-approve --id 1 --approve 1,3,5 --create-tasks

The analyzer checks every recommendation against your goals and constraints, flagging conflicts and synergies.

Agent Orchestration

Mark tasks for autonomous agent execution with configurable autonomy:

• Draft: Agent creates a draft for your review
• Checkpoint: Agent pauses at key decisions for approval
• Autonomous: Agent runs to completion

Webhooks notify your automation layer (n8n, Zapier, custom) when tasks are ready.

Playbook — Knowledge Wiki

Playbook is Quarterback's compiled knowledge layer. It's an LLM-maintained markdown wiki that gives every session — local CLI, MCP, or autonomous agent — the same canonical context about your projects, decisions, and strategies.

The problem it solves: Without Playbook, each AI session starts fresh and re-derives your organizational context from sparse signals. Two sessions running the same query can produce different results because they reconstruct understanding independently. Playbook provides accumulated knowledge that all sessions read from.

How it works:

~/playbook/                          (or ~/.quarterback/playbook/)
├── CLAUDE.md                        # Schema — how the LLM reads/writes pages
├── raw/                             # Drop zone for source material
└── wiki/
    ├── index.md                     # Master catalog — read this first
    ├── entities/                    # Companies, products, clients, tools
    ├── concepts/                    # Patterns, strategies, recurring themes
    ├── decisions/                   # Decisions with rationale and alternatives
    ├── compiled/                    # QB-compatible files for task scoring
    │   ├── goals.md                 # Read by QB's prioritization engine
    │   └── constraints.md           # Read by QB's conflict detection
    └── log.md                       # Append-only operations record

Setup: Playbook is created automatically during quarterback setup (or the MCP setup wizard). The interview asks about your key entities, concepts, and decisions, then seeds initial wiki pages.

Without Playbook: Quarterback works exactly as before — reading goals and constraints from ~/.quarterback/org-context/ files. Playbook is opt-in.

With Playbook: Quarterback reads compiled/goals.md and compiled/constraints.md from Playbook first, falling back to org-context/ files if Playbook isn't initialized. Your LLM reads wiki/index.md for full organizational context.

# Check Playbook status
quarterback playbook status

# Browse the index
quarterback playbook index

# List pages by category
quarterback playbook list --category entities

# Read a specific page
quarterback playbook read entities/my-product.md

# Search across all pages
quarterback playbook search "budget"

Obsidian integration (optional): During setup, you can opt to configure Playbook as an Obsidian vault. Open the Playbook folder in Obsidian for graph visualization and visual editing. Install an Obsidian MCP server for programmatic access. No Obsidian dependency required — Playbook works as plain markdown files.

CI/CD Pipeline Integration

Quarterback's CLI and webhook system make it a natural fit for automated pipelines — update task status, log deliverables, and trigger downstream work without a human in the loop.

Direct CLI in pipelines

Add Quarterback commands to any CI/CD step. The CLI is stateless and scriptable:

# GitHub Actions example: auto-update task on deploy
- name: Mark deploy task complete
  run: |
    pip install quarterback
    export QUARTERBACK_HOME=${{ runner.temp }}/.quarterback
    quarterback update 42 --status completed --notes "Deployed via CI, SHA: ${{ github.sha }}"

# After test suite passes, log results to a task
- name: Report test results
  run: |
    quarterback update 38 --notes "Tests passed: 106/106, coverage 87%. Build #${{ github.run_number }}"

# Nightly: check for overdue deliverables and alert
- name: Nightly priority check
  run: |
    quarterback alert-check
    quarterback priorities today --limit 5

Agentic CI/CD with webhooks

Register a webhook and let your automation layer react to task events in real time:

# Register a webhook pointing at your n8n/Zapier/custom endpoint
quarterback-server  # MCP tools available, or use CLI:

# In your automation script: mark a task agent-ready after PR merge
import subprocess
subprocess.run([
    "quarterback", "update", "55",
    "--status", "completed",
    "--notes", f"PR #{pr_number} merged. Deployed to staging."
])

Use cases: