robertsfeir/atelier-pipeline

17 stars · Last commit 2026-04-25

Multi-agent orchestration plugin for Claude Code — 10 specialized agents, persistent Brain memory, quality gates, continuous QA, wave-based parallel execution, and mechanical enforcement hooks

README preview

# Atelier Pipeline

<p align="center">
  <img src="docs/assets/logo.png" alt="Atelier Pipeline" width="480">
</p>

Multi-agent orchestration for AI-powered IDEs. Quality gates, continuous QA, and persistent institutional memory — for Claude Code and Cursor.

## What It Does

Atelier Pipeline has two core systems and two optional ones:

**Multi-Agent Orchestration.** Eleven specialized agent personas with clear responsibilities, strict boundaries, and independent quality verification. Eva orchestrates, Robert handles product, Sable designs UX, Cal architects, Colby builds, Roz tests, Poirot blind-reviews, Agatha documents, Ellis commits, Distillator compresses, and Sentinel audits security. Custom agents can be added via agent discovery. Specs get written, designs get validated, tests get authored before code, every change passes independent wave-based QA, and nothing ships without review. Works with Claude Code and Cursor.

**Atelier Brain.** A persistent memory layer backed by PostgreSQL and vector embeddings that gives your agents institutional memory across sessions. Without it, every time you close a terminal you lose the architectural decisions that shaped your implementation, the user corrections that steered scope, the rejected alternatives that explain why you didn't go a different way, and the QA lessons that prevent recurring bugs. The brain captures all of this during pipeline runs and surfaces it automatically when agents need context. It includes write-time conflict detection, TTL-based knowledge decay, and background consolidation that synthesizes raw observations into higher-level insights. The pipeline works without the brain -- but with it, session 12 of a feature build has the same context as session 1.

> **The brain is essentially free to run.** It uses OpenRouter for embeddings (`text-embedding-3-small` at $0.02/1M tokens) and occasional conflict detection (`gpt-4o-mini`). Real-world cost: **3,500+ thoughts stored over one month of heavy daily use for $0.06 total** in OpenRouter fees. Extrapolated, that's **under $0.72/year**. Fund $1.00 on OpenRouter and you're covered for a long time.

**Darwin (optional).** A self-evolving pipeline engine that queries telemetry from the brain, evaluates agent fitness, and proposes structural improvements to the pipeline itself. Enable with `darwin_enabled: true` in pipeline config. Requires the brain.

View full repository on GitHub →