extract-feature-behavior-spec

Extract a structured behavior-spec.yaml from feature tests and source.

3|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/kapilvirenahuja/garura --skill extract-feature-behavior-spec
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: extract-feature-behavior-spec
Source: https://github.com/kapilvirenahuja/garura/tree/main/core/components/skills/extract-feature-behavior-spec
Command: npx skills add https://github.com/kapilvirenahuja/garura --skill extract-feature-behavior-spec

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Extracting enterprise-ready behavior specifications from mixed source and tests is slow and error-prone, especially when teams need deterministic outcomes and precise, line-level evidence for every rule and contract.

Core Features & Use Cases

  • Tests-first intent extraction: Reads feature tests before production source to derive Given/When/Then scenarios that reflect intended behavior.
  • Tech-aware source decoding: Dispatches stack-specific extraction to synthesized temp skills so rules, contracts, and integrations match the feature’s tech stack.
  • Evidence-grade citations: Produces a behavior-spec with cited_locations at file:line precision and surfaces test-code drift as ambiguities.
  • Single-feature spec output: Generates exactly one behaviors/{feature-id}.yaml per feature during /decode, conforming to C4a shape requirements.

Quick Start

Ask the tech-architect to run extract-feature-behavior-spec for feature MEM-F001-signup to produce behaviors/MEM-F001-signup.yaml from the feature’s file surface and tests.

Frequently Asked Questions about extract-feature-behavior-spec

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

FAQPage Schema
How do I extract behavior specifications from existing tests and source code?

Extract behavior specifications by reading feature tests first to derive Given/When/Then scenarios, then validating and extending them with stack-aware source code parsing to produce a deterministic behavior-spec.yaml file.

What is tests-first intent extraction for generating behavior specs?

Tests-first intent extraction is a method that reads feature tests before production source code to derive Given/When/Then scenarios, ensuring the generated behavior specification reflects intended behavior rather than implementation drift.

How do I generate a YAML behavior contract with evidence citations from my codebase?

Generate a YAML behavior contract by parsing your feature's file surface and tests, producing cited_locations at file:line precision to provide evidence-grade citations for every extracted business rule and data contract.

Can I use behavior spec extraction for features with conflicting test and source evidence?

Yes, behavior spec extraction handles conflicting evidence by explicitly reporting ambiguities and knowledge gaps in the output, ensuring the system never silently chooses between conflicting test and source code evidence.

Does behavior spec extraction support parsing multiple tech stacks for integration behaviors?

Yes, behavior spec extraction dispatches stack-specific source decoding by detecting your tech stack, synthesizing temporary parsing skills so extracted rules, data contracts, and integration behaviors match your specific framework.

What are the limitations of single-feature behavior spec generation?

Single-feature behavior spec generation is limited to producing exactly one behaviors/{feature-id}.yaml per feature during decoding, requiring distinct input validation and dispatching for each feature's unique file surface and tech stack.