A methodology framework for collaborating with Claude on software projects. Core rules that bind every session, detailed procedures for planning, implementation, and audits, plus project-level templates.
Enterprise Development System - Best practices from Google, Spotify, AWS Well-Architected
AI 协作开发规范 —— 通用的、项目无关的 AI 开发方法论、模板与检查清单。任何项目开工前让 AI 先读本库。
FIRVP — Federal Intelligence and Regulated-data Vulnerability Prioritization. A methodology for context-aware, audit-grade vulnerability prioritization integrating federal vulnerability intelligence and regulated-data sensitivity for U.S. organizations with compliance obligations.
Methodology case study: prompt-injection detector evaluation under cross-family OOD shift. Detectors trained on direct injection meet attack families they did not see — do they still work?
AI agent methodology navigation skill — 35-methodology index map with routing tables, invocation protocols, degradation paths, and correction mechanisms for choosing the right framework first
This repository documents my game project, examining the conceptual and technical dimensions of world-building through a moral parsimony narrative.
A parallel development methodology for Claude Code - coordinate multiple AI sessions with git worktrees
This is a methodology for Harness Engineering — the practice of wrapping an LLM with the guardrails, workflows, and knowledge infrastructure that turn it from a chat interface into a reliable AI collaborator.
Foundation Claude Code harness — subagents · skills · hooks · methodology
DevDep is a Python package that implements a structured AI development methodology with skill-driven, test-driven development gates enforced via FastAPI and OpenAI.
Methodology, rules, and templates for using Claude Code on real projects — without the assistant silently making things worse. Sibling to claude-code-guardrails.
カンファレンス企画の伴走者 / A reusable framework for conference planning with Claude Code as your companion
VOID diagnostic method, claim boundaries, and implementation principles.
A battle-tested methodology and templates for building production software with AI agents — the Conductor pattern, used to ship three production systems solo.
SSOT methodology applied to AI agent skills
Playbook for generating professional DOCX reports with python-docx: the 3 mandatory pitfalls, the 10-section audit/comparison structure, color coding, verification-first workflow, and document delivery on Telegram.
A recursive skill that stress-tests any conceptual framework through 6 rounds of adversarial refinement. Say "Execute SOP" — your framework faces 7 rounds of attack. Survive or die, you win either way. It evolves every time you use it. 执行SOP——六轮攻击，七轮精炼。零安装。零配置。活下来或被杀死，你都是赢家。每用一次，进化一次。
Showcase metodológico: framework defensivo para endpoint personal Windows 11 Home. Modelo 3 capas Prevención/Detección/Recuperación + ADRs + templates PowerShell read-only. Engineered by Kamil Slodki.
Frameworkless, buildless, single-page, micro-frontend, web component pattern
Add a description, image, and links to the methodology topic page so that developers can more easily learn about it.
To associate your repository with the methodology topic, visit your repo's landing page and select "manage topics."