Product Tracking Skill

This skill tracks the completion progress of product development with support for hierarchical product structure.

Product Structure

Hybrid Product Organization

The system supports both flat and hierarchical product structures with automatic detection:

code

Option 1: Flat Structure (Simple)
└── .claude/product/
    └── index.md                  # Single file with all features

Option 2: Hierarchical Structure (Organized)
└── .claude/product/
    ├── index.md                  # Product overview and goals
    └── domains/
        ├── authentication/
        │   ├── index.md          # Domain overview
        │   └── features/
        │       ├── login.md
        │       ├── register.md
        │       └── logout.md
        ├── user-management/
        │   ├── index.md
        │   └── features/
        │       ├── profile.md
        │       └── settings.md
        └── dashboard/
            ├── index.md
            └── features/
                └── main-dashboard.md

Auto-Detection Algorithm

bash

# Step 1: Check for hierarchical structure first
if exists(".claude/product/domains/*/index.md"):
    structure_type = "hierarchical"
    product_root = ".claude/product"
    use_domains = true
else:
    # Step 2: Fall back to flat structure
    if exists(".claude/product/index.md"):
        structure_type = "flat"
        product_root = ".claude/product"
        use_domains = false
    else:
        error("No product specification found")

Priority Order:

•Hierarchical (.claude/product/domains/*/index.md) - preferred for organized projects
•Flat (.claude/product/index.md) - simple flat format for projects without domains

Parsing Product Structure

For Hierarchical Structure

bash

# 1. Detect structure
domain_files = Glob(".claude/product/domains/*/index.md")

if domain_files.length > 0:
    # Hierarchical structure detected
    structure_type = "hierarchical"

    # 2. Read product index
    Read .claude/product/index.md
    # Extract: product name, overview, goals

    # 3. Discover all domains
    for domain_file in domain_files:
        Read domain_file
        # Extract: domain name, description, priority

        # 4. For each domain, discover features
        domain_name = extract_from_path(domain_file)
        feature_files = Glob(".claude/product/domains/{domain_name}/features/*.md")

        # 5. For each feature, read feature details
        for feature_file in feature_files:
            Read feature_file
            # Extract: feature name, priority, acceptance criteria, subtasks

For Flat Structure

bash

# 1. Detect structure
if exists(".claude/product/index.md"):
    product_file = ".claude/product/index.md"
else:
    error("No product specification found")

# 2. Read product specification
Read product_file
# Extract: product name, overview, goals, feature list

# 3. Parse features from markdown
# Features are typically under "## Features" section
# Each feature has: name, priority, acceptance criteria

State File: .agentful/completion.json

Hierarchical Schema

json

{
  "domains": {
    "domain-id": {
      "name": "Domain Name",
      "priority": "CRITICAL|HIGH|MEDIUM|LOW",
      "status": "pending|in_progress|complete|blocked",
      "score": 0-100,
      "started_at": "2026-01-18T00:00:00Z",
      "completed_at": "2026-01-18T01:00:00Z",
      "features": {
        "feature-id": {
          "name": "Feature Name",
          "priority": "CRITICAL|HIGH|MEDIUM|LOW",
          "status": "pending|in_progress|complete|blocked",
          "score": 0-100,
          "started_at": "2026-01-18T00:00:00Z",
          "completed_at": "2026-01-18T01:00:00Z",
          "subtasks": {
            "subtask-id": {
              "name": "Subtask Name",
              "status": "pending|in_progress|complete",
              "completed_at": "2026-01-18T01:00:00Z"
            }
          },
          "notes": "Optional notes about progress"
        }
      },
      "notes": "Optional notes about domain progress"
    }
  },
  "gates": {
    "tests_passing": false,
    "no_type_errors": false,
    "no_dead_code": false,
    "coverage_80": false,
    "security_clean": false
  },
  "overall": 0,
  "last_updated": "2026-01-18T00:00:00Z"
}

Flat Schema (Legacy Support)

For products without domain structure, the flat schema is still supported:

json

{
  "features": {
    "feature-id": {
      "status": "pending|in_progress|complete|blocked",
      "score": 0-100,
      "started_at": "2026-01-18T00:00:00Z",
      "completed_at": "2026-01-18T01:00:00Z",
      "notes": "Optional notes about progress"
    }
  },
  "gates": {
    "tests_passing": false,
    "no_type_errors": false,
    "no_dead_code": false,
    "coverage_80": false,
    "security_clean": false
  },
  "overall": 0,
  "last_updated": "2026-01-18T00:00:00Z"
}

Reading Progress

Hierarchical Structure

bash

# Read current state
Read .agentful/completion.json

# Display hierarchical progress
# Overall: 48%
# Domains: 2/3 complete
#   - Authentication: 100% (3/3 features)
#   - User Management: 45% (1/2 features)
#   - Dashboard: 0% (0/1 features)
# Gates passing: 3/5

Flat Structure (Legacy)

bash

# Read current state
Read .agentful/completion.json

# Parse and display
# Overall: 48%
# Features complete: 2/5
# Gates passing: 3/5

Updating Progress

Updating Hierarchical Structure

When a subtask is completed:

json

{
  "domains": {
    "authentication": {
      "features": {
        "login": {
          "subtasks": {
            "login-ui": {
              "status": "complete",
              "completed_at": "2026-01-18T01:00:00Z"
            },
            "login-api": {
              "status": "in_progress"
            }
          },
          "score": 50,
          "status": "in_progress"
        }
      },
      "score": 50,
      "status": "in_progress"
    }
  }
}

When a feature is completed (all subtasks done):

json

{
  "domains": {
    "authentication": {
      "features": {
        "login": {
          "status": "complete",
          "score": 100,
          "completed_at": "2026-01-18T02:00:00Z",
          "notes": "Login UI and API fully implemented with tests"
        }
      },
      "score": 33,
      "status": "in_progress"
    }
  }
}

When starting work on a feature:

json

{
  "domains": {
    "authentication": {
      "features": {
        "register": {
          "status": "in_progress",
          "score": 0,
          "started_at": "2026-01-18T02:30:00Z",
          "subtasks": {
            "register-ui": {
              "status": "in_progress"
            },
            "register-api": {
              "status": "pending"
            }
          }
        }
      },
      "score": 22,
      "status": "in_progress"
    }
  }
}

Updating Flat Structure (Legacy)

When work is completed on a feature:

json

{
  "features": {
    "authentication": {
      "status": "complete",
      "score": 100,
      "started_at": "2026-01-18T00:00:00Z",
      "completed_at": "2026-01-18T01:30:00Z",
      "notes": "JWT authentication fully implemented with tests"
    }
  }
}

When starting work on a feature:

json

{
  "features": {
    "user-profile": {
      "status": "in_progress",
      "score": 0,
      "started_at": "2026-01-18T01:30:00Z"
    }
  }
}

When progress is made:

json

{
  "features": {
    "user-profile": {
      "status": "in_progress",
      "score": 45,
      "notes": "Backend service complete, frontend pending"
    }
  }
}

Calculating Overall Score

Hierarchical Score Calculation

javascript

// For hierarchical structure with domains
function calculateHierarchicalScore(domains, gates) {
  let totalDomainScore = 0;
  let domainCount = 0;

  // Calculate each domain's score
  for (const [domainId, domain] of Object.entries(domains)) {
    const features = domain.features;
    if (!features || Object.keys(features).length === 0) continue;

    // Calculate feature scores (average of subtasks or direct score)
    let featureTotal = 0;
    let featureCount = 0;

    for (const [featureId, feature] of Object.entries(features)) {
      if (feature.subtasks) {
        // Calculate from subtasks
        const subtasks = Object.values(feature.subtasks);
        const completed = subtasks.filter(st => st.status === 'complete').length;
        feature.score = Math.round((completed / subtasks.length) * 100);
      }
      featureTotal += feature.score;
      featureCount++;
    }

    // Domain score is average of its features
    domain.score = Math.round(featureTotal / featureCount);
    totalDomainScore += domain.score;
    domainCount++;
  }

  // Calculate gate scores
  const gateScores = Object.values(gates).map(g => g ? 100 : 0);
  const gateScore = gateScores.reduce((a, b) => a + b, 0) / gateScores.length;

  // Overall is weighted average of domains and gates
  const domainScore = domainCount > 0 ? totalDomainScore / domainCount : 0;
  const overall = Math.round((domainScore * 0.8) + (gateScore * 0.2));

  return { overall, domainScore, gateScore };
}

Priority Weights:

Priority	Weight
CRITICAL	1.5x
HIGH	1.2x
MEDIUM	1.0x
LOW	0.5x

Weighted score calculation:

javascript

function calculateWeightedScore(domains) {
  let totalWeightedScore = 0;
  let totalWeight = 0;

  const priorityWeights = {
    CRITICAL: 1.5,
    HIGH: 1.2,
    MEDIUM: 1.0,
    LOW: 0.5
  };

  for (const [domainId, domain] of Object.entries(domains)) {
    const weight = priorityWeights[domain.priority] || 1.0;
    totalWeightedScore += domain.score * weight;
    totalWeight += weight;
  }

  return Math.round(totalWeightedScore / totalWeight);
}

Flat Score Calculation (Legacy)

javascript

// Average of all feature scores
const featureScores = Object.values(features).map(f => f.score);
const featureScore = Math.round(
  featureScores.reduce((a, b) => a + b, 0) / featureScores.length
);

// Gate scores
const gateScores = Object.values(gates).map(g => g ? 100 : 0);
const gateScore = gateScores.reduce((a, b) => a + b, 0) / gateScores.length;

// Overall is average of features and gates
const overall = Math.round((featureScore * 0.8) + (gateScore * 0.2));

// Adjusted by gates (legacy method)
const gatePenalty = Object.values(gates).filter(g => !g).length * 5;
const finalScore = Math.max(0, overall - gatePenalty);

Quality Gates

Each gate must pass for production readiness:

Gate	Check	Command
tests_passing	All tests pass	`npm test`
no_type_errors	No TypeScript errors	`npx tsc --noEmit`
no_dead_code	No unused code	`npx knip`
coverage_80	Test coverage ≥ 80%	`npm test -- --coverage`
security_clean	No secrets/vulnerabilities	`npm audit`

Feature Status Values

Status	Meaning	Score Range
pending	Not started	0
in_progress	Work in progress	1-99
complete	Fully done and validated	100
blocked	Waiting on decision/dependency	any

Integration with Product Structure

Parsing Hierarchical Product Structure

Parse .claude/product/index.md and all domain/feature files:

markdown

<!-- .claude/product/index.md -->
# My Product

## Domain: Authentication
Priority: CRITICAL

## Domain: User Management
Priority: HIGH

<!-- .claude/product/domains/authentication/index.md -->
# Authentication Domain

Provides user authentication and authorization features.

## Features
- User Registration
- User Login
- Password Reset

<!-- .claude/product/domains/authentication/features/login.md -->
# Feature: User Login

Priority: CRITICAL

## Subtasks
1. Create login form UI
   - Status: pending
   - Acceptance: [ ] Email field, [ ] Password field, [ ] Submit button
2. Implement login API
   - Status: pending
   - Acceptance: [ ] POST /api/auth/login, [ ] JWT generation, [ ] Error handling

Map to completion.json:

json

{
  "domains": {
    "authentication": {
      "name": "Authentication",
      "priority": "CRITICAL",
      "status": "in_progress",
      "score": 33,
      "features": {
        "login": {
          "name": "User Login",
          "priority": "CRITICAL",
          "status": "in_progress",
          "score": 50,
          "subtasks": {
            "login-ui": {
              "name": "Create login form UI",
              "status": "complete",
              "completed_at": "2026-01-18T01:00:00Z"
            },
            "login-api": {
              "name": "Implement login API",
              "status": "pending"
            }
          }
        }
      }
    }
  }
}

Parsing Flat Product Structure

Parse .claude/product/index.md to extract feature list:

markdown

## Features

### 1. Authentication - CRITICAL
**Description**: User login with JWT
**Acceptance**:
- [x] Login endpoint
- [x] Registration endpoint
- [x] JWT token generation
- [ ] Refresh token flow

Map to completion.json:

json

{
  "features": {
    "authentication": {
      "status": "in_progress",
      "score": 75,
      "acceptance": {
        "login_endpoint": true,
        "registration_endpoint": true,
        "jwt_generation": true,
        "refresh_token": false
      }
    }
  }
}

Usage

Auto-Detection First

Always detect structure type before any operation:

bash

# Detect structure
domains_found = Glob(".claude/product/domains/*/index.md")
product_index_exists = exists(".claude/product/index.md")

if domains_found:
    use_hierarchical_tracking()
else if product_index_exists:
    use_flat_tracking()
else:
    error("No product specification found")

For Hierarchical Structure

When orchestrator asks for progress update:

•Detect structure type (domains found → hierarchical)
•Read .claude/product/index.md to discover domains
•Use Glob to find all .claude/product/domains/*/index.md files
•For each domain, read its index and discover features
•Read .agentful/completion.json
•Calculate overall percentage with weighted hierarchy
•Identify next priority subtask within highest priority domain
•Check for blocked items
•Report summary with domain/feature/subtask breakdown

When subtask work completes:

•Update subtask status to "complete"
•Add completed_at timestamp
•Recalculate feature score (based on completed subtasks)
•Recalculate domain score (based on feature scores)
•Recalculate overall score (weighted average)
•Write back to completion.json

For Flat Structure

When orchestrator asks for progress update:

•Detect structure type (no domains, .claude/product/index.md exists → flat)
•Read .claude/product/index.md for product specification
•Read completion.json
•Calculate overall percentage
•Identify next priority feature
•Check for blocked items
•Report summary

When feature work completes:

•Update feature status to "complete"
•Set score to 100
•Add completed_at timestamp
•Recalculate overall score
•Write back to completion.json

Structure Compatibility

Structure Type	Detection	Product File	Completion Schema	Tracking Method
Hierarchical	`.claude/product/domains/*/index.md` exists	`.claude/product/index.md`	Nested `domains` object	Track at subtask → feature → domain levels
Flat	`.claude/product/index.md` exists	`.claude/product/index.md`	Flat `features` object	Track at feature level