functional-spec

Generates evidence-tagged functional specifications with MVP modules, use cases, business rules, and Firestore data models.

Updated Jun 28, 2026
One-click install
npx skills add https://github.com/JaviMontano/claude-plugins --skill functional-spec-javimontano
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: functional-spec
Source: https://github.com/JaviMontano/claude-plugins/tree/main/plugins/claude-native-toolkit/skills/functional-spec
Command: npx skills add https://github.com/JaviMontano/claude-plugins --skill functional-spec-javimontano

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Writing functional specifications manually often produces vague feature lists, untestable requirements, and missing acceptance criteria. This Skill produces structured, evidence-tagged functional specs with enforced completeness rules before architecture or development begins. ## Core Features & Use Cases - Deterministic Spec Compilation: Validates and renders structured JSON specs into Markdown via scripts/compile-functional-spec.py, rejecting specs with fewer than 8 use cases or missing acceptance criteria. - Evidence-Tagged Analysis: Every finding is tagged [CODE], [CONFIG], [DOC], [INFERENCE], or [ASSUMPTION], with a warning banner when assumptions exceed 30%. - Platform Feasibility Lens: Maps requirements to Firebase (Firestore), Vercel, and Hostinger platforms, including Firestore collection metadata with owner, PII, indexes, and retention. - Use Case: Given stakeholder interviews and policy documents for a field-service MVP, produce a spec with MVP modules, 8+ actor-goal use cases, typed business rules traceable to use cases, and Given/When/Then acceptance criteria. ## Quick Start Ask the AI to create a functional specification for your MVP from your project documents, including at least eight use cases, business rules, acceptance criteria, and Firestore data model notes.

Frequently Asked Questions about functional-spec

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a functional specification for an MVP?▼

Define MVP modules, then write at least eight actor-goal use cases with triggers, preconditions, main flows, and acceptance criteria. Add typed business rules linked to use cases, data model notes, out-of-scope boundaries, and open questions.

How to validate a functional spec automatically with a script?▼

Express requirements as structured JSON and run compile-functional-spec.py with the --spec flag. The script validates use case count, ID patterns, business rule taxonomy, and Firestore collection metadata, failing nonzero on incomplete specs.

What business rule types should a functional spec include?▼

The supported taxonomy covers eligibility, validation, workflow, authorization, notification, data-retention, billing, and audit rules. Each rule needs an ID matching BR-###, a statement, a source, and links to known use cases.

Does the functional spec include implementation code for Firebase?▼

No. The spec stays in analysis mode and records Firestore feasibility as collection-level notes with owner, PII, fields, indexes, and retention. Implementation details are deferred to later architecture and delivery phases.

What happens when requirements are mostly unknown or assumed?▼

Unsupported statements are tagged as assumptions and the evidence-tag distribution is reported. When assumptions exceed 30 percent, a prominent warning banner is added, and unresolved items are captured as open questions rather than invented requirements.