Extract evidence-based EARS-formatted requirements from legacy codebases.
npx skills add https://github.com/jsdfhasuh/emosonic --skill spec-miner-jsdfhasuh
Please help me install this Agent Skill. Skill: spec-miner Source: https://github.com/jsdfhasuh/emosonic/tree/main/.agents/skills/spec-miner Command: npx skills add https://github.com/jsdfhasuh/emosonic --skill spec-miner-jsdfhasuh
Spec miner turns legacy, undocumented, or opaque implementations into clear, evidence-based requirements so teams can safely understand, document, and improve existing systems.
Use the spec-miner skill to reverse-engineer the target repository and produce an evidence-backed requirements document that includes EARS-formatted functional observations and a list of uncertainties.
High-intent search queries and answers about installing and using this skill.
Reverse engineering legacy code into specifications grounds observations in actual code evidence and records locations for each requirement. This process traces system behavior to document observed functionality without relying on assumptions.
EARS (Easy Approach to Requirements Syntax) is a structured syntax for drafting consistent requirements. It formats functional observations extracted from code into a standardized style suitable for downstream planning and validation.
Documenting observed system behavior for onboarding involves tracing execution flows and converting scattered behaviors into navigable specifications. This creates explicit documentation with flagged uncertainties and recommendations.
Spec extraction works for undocumented implementations by following a structured reverse-engineering workflow across scope discovery. It traces flows to produce evidence-backed functional observations and lists identified uncertainties.
The best way to handle uncertainty when reverse engineering code is to explicitly flag ambiguities encountered during flow tracing. This records gaps in observed behavior alongside generated specifications and recommendations.
remiverdiesenClone Ark's repository and inspect its architecture, CRDs, and controllers.
windyskyOrchestrate SPEC workflows with EARS format and Plan-Run-Sync lifecycle.
juanjaragaviLocate and map code implementations, dependencies, and relationships across large codebases.
zenobi-usAutomate API documentation creation and maintenance with OpenAPI integration.
