adlc

On the path to fully autonomous agentic engineering

Agentic Development Lifecycle

Local AI-powered code review agents for Claude Code

Evaluate (lint, review, and security-audit) your AI coding agent harness.

Generative Architectural Brain Base Engine - Agentic Software R&D Engineering Kit - development, all languages, agents, platforms, with Engineering Team & Experimental Cognitive Brain, Software Engineering & Architecture Best Practices and Governance

Open-source Claude Code plugin: AI delivery pipeline — plan epics/stories/tasks, build with verified checks, production-safe QA, docs. Azure DevOps · Jira · GitHub · Teams · Slack

A spec-driven pipeline that makes building with AI fast, accountable, and cumulative.

Plan the work. Control the context. Ship with proof.

Improving DX for Analytics Engineers

A Claude Code mod that summarizes the agent messages you have not read, in plain English. Run /catchup, type 'brief me', or press a button.

🥇 Terraform Registry-publishable repo with Agent Development Lifecycle (ADLC) governance, 3-tier testing, and FOCUS 1.2+ FinOps compliance 💎

Self-hosted toolkit for teams maintaining legacy codebases

An Agentic Development Lifecycle (ADLC) protocol for AI coding agents.

The operating model for agentic development: 47 skills, 28 commands and 2 agents for readiness, agent-ready specs, review capacity, governance and evals. Claude Code, Codex & Cursor.

Dual-mode CLI for coding agents: skill management + agent execution + team lifecycle (setup, update, repair). Install skills, generate slash commands, wire lifecycle events, run any agent headlessly.

Agentic Development Lifecycle skills for real software work

A security gate for AI agents: reads an agent's package before install and answers passed, refused or declined, checking what the code reaches against what the package declares. Implements the ADLC security reviewer role.

The Agentic Development Lifecycle end to end: build software with coding agents — spec → plan → implement → verify → secure → release — with humans owning every gate. Claude Code + OpenSpec + Cloudflare free tier.

Yadflow (yad = يد, "hand") — the governance layer for AI-assisted development: AI builds, a human hand clears every gate. Gated, team, multi-repo author → review → build, with a PR-driven review gate, CI gates and a zero-dependency `yad` CLI. Agent-agnostic: 38 skills run in Claude Code, Codex, Cursor, Copilot, Gemini CLI and more.

A lightweight Python CLI tool that orchestrates an AI coding agent as part of an AI-assisted Development Loop (ADLC) on a legacy system.