DYAI2025/Plumbline

6 stars · Last commit 2026-06-08

Plumbline — a self-learning, customer-value-governed agile AI agent team for Claude Code. 87 subagents + skills, TDD defense-in-depth gates, Kaizen retros, a four-body adversarial council, and an empirically benchmarked QA harness. Does it hang true?

README preview

<div align="center">

# Plumbline

### *Does it hang true?*

**A defense-in-depth agent framework for Claude Code — built around one obsession: proving that work is *actually* done, not that it merely *looks* done.**

`86 subagents` · `16 vendored skills` · `/agileteam` v3 orchestrator · `/concilium` four-body council · `Reality-Ledger QA` · `empirically benchmarked`

**An agile AI agent framework for Claude Code: a self-learning, customer-value-governed agentic team that builds software with TDD gates, Kaizen retrospectives, and a defense-in-depth quality pipeline.**

`#ClaudeCode` `#ClaudeSkills` `#AgentFramework` `#AIAgents` `#AgenticAI` `#AgileAIAgents` `#AIAgile` `#AIAgenticAgileTeam` `#KaizenAgentic` `#SelfLearningAgileTeam` `#MultiAgentSystems` `#AgentEngineering` `#TDD` `#AutonomousCoding` `#DefenseInDepth` `#LLMOps`

> ### We benchmarked our own agent framework — and discovered our cleverest idea didn't work.
> Then we shipped the honest result anyway. **That** is Plumbline.

**[▶ Live demo](https://dyai2025.github.io/Plumbline/)** · explore all 86 agents in your browser, nothing to install

[![Sponsor Plumbline — fund the token costs of the benchmarks](https://img.shields.io/badge/Sponsor-fund%20the%20benchmarks-ff5fa2?logo=githubsponsors&logoColor=white)](https://github.com/sponsors/DYAI2025)

View full repository on GitHub →