GSD (Get Shit Done) is a mature, actively maintained meta-prompting and context-engineering system that orchestrates AI coding agents (Claude Code, OpenCode, Gemini, Codex, Copilot, Cursor, and 9+ other runtimes) through a structured milestone → phase → plan → execute → verify workflow. The system solves context decay in long AI-assisted coding sessions by maintaining a `.planning/` artifact tree (ROADMAP.md, STATE.md, PLAN.md, SUMMARY.md, VERIFICATION.md) and dispatching focused subagents with fresh 200K-token contexts for each discrete task. The codebase is a dual-package npm monorepo: a legacy CJS CLI (`get-shit-done-cc`) and a modern TypeScript SDK (`@gsd-build/sdk`) with a query-registry architecture, runtime-bridge sync primitive, and generated CJS mirrors that keep both surfaces in parity. The architecture is built around a 'thin orchestrator' pattern where slash-command entry points delegate to workflow markdown files that spawn specialized subagents (planner, executor, verifier, researcher, debugger, code-reviewer) via a registry-backed `gsd-sdk query` dispatch layer. The SDK exposes a `GSD` class that composes plan parsing, config loading, prompt building, and session running into `executePlan()` and `runPhase()` calls, with event-stream telemetry and CLI/WS transports. A `synckit`-based synchronous bridge (`executeForCjs`) lets CJS callers invoke async SDK handlers without subprocess overhead. The query registry resolves commands through a multi-strategy topology (exact, alias, dotted, spaced) with native handlers and transparent CJS fallback. What distinguishes this codebase is the obsessive engineering around installer migrations (transactional, with rollback and user-prompt resolution), cross-runtime path projection (PowerShell/cmd/Git-Bash/POSIX), adversarial parser testing (hostile frontmatter/roadmap fixtures), atomic state writes with file locking, prompt-injection scanning hooks, and a changeset-fragment workflow that eliminates CHANGELOG merge conflicts. The test suite is exceptionally thorough — 400+ test files covering mutation-resistant config validation, golden parity between CJS and SDK, security prompt injection, worktree safety, and regression coverage for 3800+ tracked issues. The system supports 15 AI runtimes with runtime-aware artifact layout, model-profile resolution, and per-agent model overrides.
Microsoft Semantic Kernel is an AI orchestration SDK for building multi-agent systems, process workflows, and LLM-powered applications. It provides a unified abstraction layer over multiple AI providers (OpenAI, Azure, Google, Mistral, Ollama, HuggingFace, etc.) and supports function calling, plugin systems, prompt templates, memory/vector stores, and content safety filters. The framework is implemented in both Python and .NET with consistent architectural patterns. In production, the system acts as a library that developers embed into their applications. The central Kernel class manages service registration, plugin loading, and function execution. Agents (ChatCompletionAgent, AzureAIAgent, etc.) provide conversational interfaces with tool-calling capabilities. The process framework enables event-driven, multi-step workflows with state persistence. The filter system allows cross-cutting concerns like logging, moderation, and quality checks to be injected into function invocation and prompt rendering pipelines. Key entry points include the Kernel class, agent classes, ProcessBuilder for workflows, and various sample scripts. Data flows typically involve user input being processed by an agent or kernel, which invokes AI services via connectors, optionally calls plugins/functions, and returns responses. The framework relies on external AI services, vector databases, and optionally Dapr for distributed process execution. While primarily an application framework, Semantic Kernel includes data/research-oriented components such as vector store integrations, embedding generation services, and quality evaluation filters, making it a hybrid tool suitable for both production AI applications and experimental research.
OpenMontage is an open-source, agentic video production platform where an AI coding assistant (Claude Code, Cursor) orchestrates the entire pipeline. It provides a library of 100+ tools (video generation, audio, graphics, enhancement, analysis, avatar, subtitles) behind a scored provider selection layer, and 10+ production pipeline manifests (animated explainer, cinematic, documentary montage, talking head, etc.) that define stage order, approval gates, and tool assignments. The core architecture is a Python framework with a BaseTool class, ToolRegistry for auto-discovery, pipeline loader, checkpoint system, and a Remotion-based video composer. It also includes a Backlot living storyboard server for live production monitoring and an Ink Theater procedural puppetry engine. While primarily an application, it contains data/research elements such as ManimGL/ManimCE educational animation examples covering transformer internals, quantum mechanics, and ML concepts, as well as a CLIP-based corpus builder for documentary montage pipelines.
Liza is a disciplined multi-agent orchestration system for AI-assisted software development. It manages a pipeline of roles (orchestrator, coder, reviewer, architect) that collaborate via a behavioral contract enforced by deterministic Go code. The system provides a CLI (Cobra) and a Bubbletea TUI, with a YAML-based state file (blackboard) for persistence, Git worktree isolation per task, and a state-machine task lifecycle (planning → implementation → review → merge). It integrates with multiple LLM providers (Claude, Codex, Gemini, etc.) via their CLIs or the Agent Context Protocol (ACP). The architecture follows a clean layered design: `cmd/liza/` contains CLI commands, `internal/commands/` are thin adapters, `internal/ops/` holds core business logic (claiming, submitting, reviewing, merging, recovering), `internal/models/` defines domain types, and `internal/db/` manages the blackboard. The system is heavily test-covered with unit, integration, and lifecycle replay tests. It also includes optional codebase indexing via SCIP, Stacklit, and Semble for improved context. Key features include generation-fenced agent authority, circuit breaker anomaly detection, sprint metrics, checkpoint summaries, automated dependency repair, and pre-commit hooks. The repository is well-documented with extensive ADRs, architecture diagrams, and skill definitions. While the codebase is large and has some platform-specific code (Windows support), it demonstrates high craftsmanship in its use of idempotent lifecycle receipts, file-locking primitives, and error classification.
Commonly is an open-source multi-agent collaboration platform where humans and AI agents work together in shared workspaces called 'pods'. The system provides real-time messaging, task management, agent runtime (native, hosted, or bring-your-own), marketplace for agent skills, and integrations with external chat platforms like Discord, Slack, Telegram, and GroupMe. The architecture is a full-stack monolith with separate external agent bridge services and a Cloudflare Workers-based agent runtime. The backend (Node.js/Express) uses a dual-database design (MongoDB for primary data, PostgreSQL for messages/tasks) with automatic sync. The frontend is a React SPA with a v2 design system. The CLI allows local agents to participate via device-code auth and polling. The MCP server provides standard tools for agents to interact with the platform. The codebase is well-structured with extensive test coverage (unit, integration, E2E), Architecture Decision Records (ADRs), and reusable packages for types, integration SDK, and MCP tools. Notable craftsmanship includes pluggable storage drivers, deterministic agent wake targeting, and a test utility that replicates the production PostgreSQL schema to catch DDL drift.
The google/adk-go repository is a production-grade, open-source Go framework for building, orchestrating, and deploying multi-agent AI systems. It provides a comprehensive set of abstractions including agents, workflows, tool integrations, model backends (Gemini, OpenAI, Apigee), session management, artifact storage, and human-in-the-loop (HITL) support. The architecture centers around a workflow engine that supports graph-based execution with nodes, edges, retry, concurrency, and dynamic branching. It also implements agent-to-agent (A2A) and Model Context Protocol (MCP) for inter-agent communication. The framework is designed to be deployed via REST API, WebSocket, SSE, or as a console application, and includes launchers for Google Cloud Run and Vertex AI Agent Engine. While primarily an application framework, it also includes telemetry and analytics plugins for observability. The codebase is well-structured with extensive unit and integration tests, but some internal packages and experimental features (e.g., OpenAI model adapter) indicate ongoing development and potential for interface drift.
SuperAgentX is a modular, open-source multi-agent AI framework designed for orchestrating autonomous agents that plan, act, and execute real-world workflows. It supports 100+ LLM providers (OpenAI, Azure, Gemini, Claude, Bedrock, Ollama, LiteLLM) and 10,000+ MCP tools via the Model Context Protocol. The architecture centers on `AgentXPipe` pipelines that sequence or parallelize agents, each composed of an `Engine` that wraps an LLM client and a set of tool handlers. Handlers are defined using a `@tool` decorator pattern that auto-generates LLM function schemas. The framework includes built-in human-in-the-loop approval channels (console, Slack, Teams), persistent memory with SQLite and vector stores (Chroma, OpenSearch), full observability via OpenTelemetry and SQL-based telemetry, and a CLI for project scaffolding. It also features a production-grade browser automation engine based on Playwright for tasks where APIs are unavailable.
callx is a Node.js CLI tool that serves as a unified runtime switcher and configuration manager for AI coding assistants like Claude Code and OpenCode. It allows users to define multiple AI 'sprites' (agent configurations) with different providers, models, skills, and MCP servers, all managed through a single `~/.config/callx` directory. The system provides two entry points: the `call` command for quick invocation of sprites, and `callcli`, a full-featured TUI built with Ink and React for managing providers, sprites, skills, MCP servers, and git worktrees. The architecture is a monolith CLI application with a clear separation between runtime utilities (`lib/runtime.js`), git worktree helpers (`lib/worktree.js`), and the TUI frontend (`lib/callcli/app.mjs`). The runtime module handles all configuration file I/O, provider normalization, skill syncing via symlinks, and MCP server configuration generation for both Claude Code and OpenCode formats. The TUI provides a Monokai-themed interface with list-based navigation, field-level editing, and multi-select for skills and MCP servers. Key data flows include: reading `providers.json` for API keys and endpoints, reading sprite `config.json` for CLI type and model selection, syncing skill packs from `skillpacks/` to sprite-specific `skills/` directories via symlinks, and generating MCP configuration files from `mcpservers/` definitions. The tool also integrates with git worktrees, providing an interactive flow to create new branches and worktrees with suggested naming conventions. External dependencies are minimal: Node.js 18+, Ink and React for the TUI, and the target AI CLI tools (opencode-ai, @anthropic-ai/claude-code).