software-methodology

Disciplined AI Collaboration: a method for building software with AI. One loop at every size, rules held by checks rather than attention, state derived rather than written, evidence in place of claims. Includes the PAG grammar, an architecture and a queryable ontology. Every page is also served as Markdown and JSON at banes-lab.com.

SENAR — открытый стандарт разработки с участием ИИ: правила, quality gates, метрики и чек-листы верификации. Отвечает на вопрос, при каких условиях работу можно считать сделанной. Эталонная реализация — TAUSIK.

Una aplicación interactiva para visualizar y trabajar con los parámetros del modelo Boehm-Turner.

Vision-Driven Development / Release-Driven Development framework for AI-autonomous software development with enforced guardrails.

You lead. Agents hunt. Cognitive + methodological harness for AI-assisted software — 139 skills across 7 companion packs + composition layer. Cross-tool (Claude Code, Cursor, Codex, Gemini CLI).

Software Methodology Project 2

Agent 时代的软件项目搭建方法论 | A bilingual course on decision records, mechanical gates, and plugin architecture, distilled from deepseek-harness

Reconciled Spec-Driven Development — a reasoning cascade that keeps spec and code aligned. Extends github/spec-kit with the artifact↔code reconciliation arc.

This repository contains the whitepaper "Trained Monkeys: A Methodology for AI-Orchestrated Software Development," a formal, descriptive technical report for practitioners. The paper presents a three-stage AI orchestration pipeline (planner -> decomposer -> supervisor) and argues four core properties grounded in the skill definitions.

An open source framework for AI-first engineering teams, born at Disruptica. Requests over roadmaps, contracts over tickets, agents that write the code. Companion tool: BugLady.ai / Ramjet

A working method for shipping software with coding agents, plus the linter that makes it enforceable.

Instrumented AI-assisted development for business operators and hobbyists — governance for directing an AI coding agent by dollars, by time, or both.

Overview of Software Development Life Cycle (SDLC) models and methodologies, including traditional and modern approaches with diagrams and presentation material (Software Quality & Reliability, UNIWA).

A repeatable operating system for shipping with AI agents: Context → Requirements → Shaping → Sprint Plan → Execution. Method, deck, article, and a 30s overview.

Caso de uso real: planejar e executar um app fullstack com Claude Code usando documentação viva como fonte única de verdade.

RENAR — стандарт инженерии требований для AI-нативной разработки | Requirements Engineering & Normative Adaptive Regulation (RU/EN, CC BY-SA 4.0)

코딩 에이전트용 개발방법론 플러그인. 도메인 결정을 사용자와 확정하고 필요한 설계·구현·검증을 이어간다.

Bi-directional traceability for AI-assisted development — prove every commit traces to the business goal that authorised it. 8-phase chain, 7 gates, 113 checks. MCP server at vdd.simonmak.com. Skill for Claude Code, Cursor and OpenCode.

An agent's claim that it finished is worthless unless the proof could have failed. A one-page method, no framework, no install.