A portable, dynamic multi-agents workflow system for complex codebases. Reads cc-switch config, selects the right agent architecture (loop / orchestrator-workers / multi-agent / graph / dynamic / ultracode), generates per-agent configs, enforces cost-rate

Reusable Cloudflare Worker library (Hono + D1) exposing the Agent Architecture Board core: the `createApp()` Hono factory, the default fetch/scheduled/queue handler, and the Durable Object + Workflow classes. Deployments (see deploy/backend) supply the wr

Opt-in Cloudflare Workers AI model registry for the Agent Architecture Board's AI provisioning facade. Mix into a CompositeModelProvider to add the `workers-ai` provider (in-process binding on the Worker, OpenAI-compatible REST elsewhere).

Bootstrap CLI for the Agent Architecture Board: scaffold a local-mode deployment (Node/local backend + frontend SPA) on your own machine — generates the crypto secrets, populates and gitignores the .env files, and mints a GitHub/GitLab personal access tok

The app-level caching seam for the Agent Architecture Board (docs/initiatives/caching-layer.md): createAppCaches builds the named, typed in-memory read-through caches (layered-loader) the services consume via the kernel AppCaches port, with optional distr

Versatile autonomous AI agent with three-agent architecture (Manager, Worker, Client) powered by gpt-oss-120b. Adaptive validation, full MCP support, and intelligent quality control.

The built-in polling-gate suite (CI, merge-conflicts, post-release health + the on-call escalation) for the Agent Architecture Board, authored entirely through the public gate-registry seam — depends only on @cat-factory/kernel + @cat-factory/contracts, n

Local-mode runtime facade for the Agent Architecture Board: the @cat-factory/node-server stack with agent jobs run as local Docker/Podman containers and GitHub accessed via a personal access token — a developer can run the whole product on their own machi

Phalanx multi-agent architecture for Pi — shared memory via agora, role dispatch with chain-of-command, shield_wall retry, and consult-the-oracle escalation

Neuro-Symbolic AI Agent Architecture for Claude Code

Multi-Agent Research MCP server - fully compliant with Anthropic's multi-agent architecture

Agent-driven daily brief CLI for manually curated Agent architecture and AI Coding sources.

The Salesforce CLI

Phalanx multi-agent architecture for Pi — shared memory via agora, role dispatch with chain-of-command, shield_wall retry, and consult-the-oracle escalation

Abstract filesystem capability seam (ctx.fs) for the DeepSeek Harness — vocabulary types, the FileSystem service (text IO + optional version-guarded atomic mutations), and the fs/* policy event vocabulary

AI operator scaffold for OpenClaw — identity, memory, instincts, cron, and agent architecture in one install

A module for generating metrics from V8.

A2UI Core Library

Benchmark framework for evaluating AI coding agent architecture patterns

Production-grade AI agent architecture skill with templates for coding, research, task, workflow, and API agents.