Release Notes: v0.3.0¶

Release Date: 2026-01-26

Overview¶

This release adds integration with structured-evaluation for standardized PRD quality reports, a new merge command for combining multiple JSON files, and enhanced user story fields.

Highlights¶

Structured Evaluation Integration: Convert deterministic scoring results to EvaluationReport format for consistent output across LLM-based and rule-based evaluations
Merge Command: New CLI command to combine multiple PRD JSON files with deep merge support
Enhanced User Stories: Additional fields for better Agile workflow integration

New Features¶

Evaluation Integration¶

The new evaluation integration enables PRD scoring results to be output in the standardized EvaluationReport format from structured-evaluation:

import "github.com/grokify/structured-plan/prd"

// Convert deterministic scoring to EvaluationReport
doc, _ := prd.Load("my-product.prd.json")
report := prd.ScoreToEvaluationReport(doc, "my-product.prd.json")

// Or generate a template for LLM judge evaluation
template := prd.GenerateEvaluationTemplate(doc, "my-product.prd.json")

Functions Added:

Function	Description
`ScoreToEvaluationReport()`	Converts deterministic scoring results to EvaluationReport
`GenerateEvaluationTemplate()`	Creates empty template for LLM judge to fill in
`GenerateEvaluationTemplateWithWeights()`	Template with custom category weights
`StandardCategories()`	Returns 10 standard PRD evaluation categories
`CategoryDescriptions()`	Category ID to description map for LLM prompts
`CategoryOwners()`	Category ID to suggested owner map
`GetCategoriesFromDocument()`	Extracts categories including custom sections

Standard Categories (10 total):

Category	Weight	Owner
problem_definition	20%	problem-discovery
solution_fit	15%	solution-ideation
user_understanding	10%	user-research
market_awareness	10%	market-intel
scope_discipline	10%	prd-lead
requirements_quality	10%	requirements
metrics_quality	10%	metrics-success
ux_coverage	5%	ux-journey
technical_feasibility	5%	tech-feasibility
risk_management	5%	risk-compliance

Merge Command¶

Combine multiple PRD JSON files into one with deep merge support:

# Merge multiple files
srequirements merge base.json overlay.json -o combined.json

# Merge with default output (merged.json)
srequirements merge part1.json part2.json part3.json

Deep merge behavior: - Nested objects are recursively merged - Later files override earlier values for scalar fields - Arrays are replaced (not concatenated)

Enhanced User Stories¶

User stories now support additional fields for Agile workflows:

{
  "id": "US-001",
  "title": "User Authentication",
  "epic_id": "EPIC-AUTH",
  "story_points": 5,
  "labels": ["security", "mvp"],
  "dependencies": ["US-002", "US-003"]
}

Dependencies¶

Updated structured-evaluation to v0.2.0 (adds rubrics, judge metadata, pairwise comparison)

Migration¶

No breaking changes. Existing code continues to work without modification.

Installation¶

go get github.com/grokify/structured-plan@v0.3.0

Or install the CLI:

go install github.com/grokify/structured-plan/cmd/srequirements@v0.3.0