generate-functional-from-code

Generate a functional graph from a codebase using the legacy cluster pipeline.

2|2|Updated Mar 11, 2026
One-click install
npx skills add https://github.com/accionlabs/breezeai-claude-plugin --skill generate-functional-from-code
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: generate-functional-from-code
Source: https://github.com/accionlabs/breezeai-claude-plugin/tree/main/skills/generate-functional-from-code
Command: npx skills add https://github.com/accionlabs/breezeai-claude-plugin --skill generate-functional-from-code

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires boto3, requests, numpy, scikit-learn.

What problem does it solve?

It creates a functional graph (Persona → Outcome → Scenario → Step → Action) from a codebase when no UI exists and a fast, batched pipeline is needed, avoiding the complexity of split UI/back‑end pipelines.

Core Features & Use Cases

  • Legacy Cluster Pipeline: Runs a multi‑pass Python pipeline that extracts intents, deduplicates with DBSCAN, assigns outcomes, and generates scenarios.
  • Fast Generation: Ideal for backend‑only repositories or projects where a quick functional model is preferred over detailed UI‑driven analysis.
  • Fallback Capability: Serves as a reliable fallback when the newer split pipelines cannot be applied.

Quick Start

Invoke the skill with the command to generate a functional graph from your repository.

Frequently Asked Questions about generate-functional-from-code

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

FAQPage Schema
How do I generate a functional graph from a backend codebase?

A functional graph represents code clusters as a structured hierarchy: Persona, Outcome, Scenario, Step, and Action. It maps backend-only repository code ontology into a functional model using batched DBSCAN clustering to deduplicate intents and assign outcomes.

Can I use DBSCAN clustering to map code ontology into functional scenarios?

Yes, DBSCAN clustering is used to deduplicate code ontology clusters and extract intents during the multi-pass pipeline. This batched approach rapidly generates functional scenarios for backend repositories lacking UI components.

Do I need a Breeze project setup to run the legacy cluster pipeline?

Yes, the legacy cluster pipeline requires a Breeze project setup with a valid API key, AWS credentials, and uploaded code ontology clusters to execute the batched functional graph generation process.

When should I use the legacy cluster pipeline instead of split UI and backend pipelines?

Use the legacy cluster pipeline for backend-only repositories lacking UI where fast batched generation is acceptable. It serves as a reliable fallback when split UI and backend pipelines cannot be applied or are unnecessarily complex.

What are the limitations of generating a functional graph from code clusters?

Limitations include applicability to backend-only repositories lacking UI components, where fast batched DBSCAN generation is acceptable. It prioritizes speed over detailed UI-driven analysis and serves as a fallback when newer split pipelines are not applicable.