draft-reference-algorithms

Generate language-agnostic reference pseudocode for complex internal interfaces from tech.yaml.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Complex internal interfaces often lack clear, language-agnostic algorithm guidance, making implementations inconsistent, hard to validate, and difficult to maintain across teams.

Core Features & Use Cases

  • Language-agnostic reference pseudocode: Produces one documented algorithm section per qualifying internal interface ID to reduce ambiguity during implementation.
  • Deterministic artifact generation: Writes a single output file reference-algorithms.md with a DRAFT status and respects LOCKED artifacts to avoid overwriting.
  • Complexity-aware guidance: Classifies interfaces into complexity classes (state-machine, scheduler, retry-backoff, reconciliation, other) and extracts invariants and edge cases from behavior_contracts.
  • Use Case: When you have an AI-native SDLC design step and a tech.yaml describing internal interfaces, this skill generates consistent reference algorithm sections to speed up build and review.

Quick Start

Ask the system to run draft-reference-algorithms with interface_ids and the tech_yaml_path pointing to the relevant issue design/tech.yaml to generate reference-algorithms.md.

Frequently Asked Questions about draft-reference-algorithms

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

FAQPage Schema
How do I generate reference pseudocode from tech.yaml interface contracts?

To generate reference pseudocode from tech.yaml, run the draft-reference-algorithms skill with the interface_ids and tech_yaml_path. It reads behavior_contracts to extract invariants and outputs a reference-algorithms.md file.

What is deterministic algorithm reference generation for internal interfaces?

Deterministic algorithm reference generation produces language-agnostic pseudocode for complex internal interfaces to reduce implementation ambiguity. It classifies complexity into state-machine, scheduler, retry-backoff, or reconciliation categories to ensure consistent builds.

Can I generate interface pseudocode without modifying the original tech.yaml file?

Yes, you can generate interface pseudocode without modifying tech.yaml. The skill operates with read-only behavior on the tech.yaml file, extracting behavior_contracts to write a separate reference-algorithms.md artifact.

Does generating reference algorithms overwrite locked design documentation?

No, generating reference algorithms does not overwrite locked documentation. The skill writes a reference-algorithms.md file with a DRAFT status and explicitly respects LOCKED artifacts to prevent overwriting existing design guidance.

How do I classify interface complexity when extracting invariants and edge cases?

Interface complexity is classified automatically by analyzing behavior_contracts into state-machine, scheduler, retry-backoff, reconciliation, or other categories. This classification drives the extraction of invariants and edge cases for the pseudocode output.

What is the best way to document algorithmic interface contracts for implementation readiness?

The best way to document algorithmic interface contracts is generating language-agnostic reference pseudocode sections per interface ID. This provides deterministic, implementation-ready guidance that speeds up build and review processes across teams.