spec-miner

Extract evidence-backed specifications from legacy codebases into structured documents.

14|Updated Feb 5, 2026
One-click install
npx skills add https://github.com/alexander-danilenko/ai-skills --skill spec-miner-alexander-danilenko
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec-miner
Source: https://github.com/alexander-danilenko/ai-skills/tree/main/skills/spec-miner
Command: npx skills add https://github.com/alexander-danilenko/ai-skills --skill spec-miner-alexander-danilenko

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Helps teams understand legacy or undocumented systems by extracting and documenting specifications directly from existing code.

Core Features & Use Cases

  • Reverse-engineering to surface requirements and constraints from codebases.
  • Produces EARS-formatted requirements and a formal reverse-spec document.
  • Annotates uncertainties and preserves exact code locations as evidence.

Quick Start

Analyze a legacy codebase to generate a complete reverse-engineered specification document saved to specs/{project_name}_reverse_spec.md.

Frequently Asked Questions about spec-miner

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

FAQPage Schema
How do I extract requirements from an undocumented legacy codebase?

Extracting specifications from a legacy codebase involves reverse-engineering existing code to produce a formal specification document. The process surfaces observed requirements, architecture notes, and uncertainties as evidence-backed EARS-formatted statements.

What is EARS formatting for reverse-engineered requirements?

EARS formatting is a structured method for writing formal requirements. When applied to reverse-engineering legacy code, it transforms observed code behaviors into standardized, evidence-backed requirement statements that include exact code locations as proof.

How do I generate documentation for an unfamiliar codebase during onboarding?

Generating documentation for an unfamiliar codebase during onboarding requires code archaeology to extract formal specifications. This produces a structured reverse-spec markdown document detailing observed requirements, architecture notes, and uncertainties.

Does spec-miner require external dependencies to analyze legacy code?

Spec-miner requires no external dependencies to analyze legacy code. It operates by enforcing evidence-based observation using built-in file reading, searching, and bash execution tools to extract and document specifications directly from the codebase.

What is the best way to document uncertainties when reverse-engineering specifications?

The best way to document uncertainties when reverse-engineering specifications is to annotate them within a formal reverse-spec document. This approach preserves exact code locations as evidence alongside the extracted EARS-formatted requirements to highlight unknowns.