Opencast is a mature, open-source lecture capture and video management platform designed for educational institutions. It provides a complete pipeline for recording, processing, managing, and distributing academic video content. The system is built as a modular Java application using OSGi (Karaf) for service composition, with a REST API and a GraphQL layer for modern client interactions. Core capabilities include scheduled capture from lecture hall devices, automated workflow-driven processing (transcoding, transcription, captioning, silence detection, video editing), versioned asset storage (with AWS S3 support), multi-channel distribution (YouTube, OAI-PMH, streaming servers), and a rich admin UI for metadata management, access control, and analytics. The architecture follows a microservices-like pattern with separate API, implementation, and remote proxy modules, enabling distributed deployment across worker, admin, and presentation nodes. External dependencies include Elasticsearch/OpenSearch for search indexing, a relational database (MySQL/MariaDB/PostgreSQL), FFmpeg for media inspection and processing, and various cloud services for transcription and publication. The codebase demonstrates strong engineering practices with consistent module structure, extensive test coverage, and a well-documented upgrade path spanning over a decade of development.
ChatDev 2.0 (DevAll) is a zero-code multi-agent orchestration platform for building and executing graph-based AI workflows. Users define directed acyclic or cyclic graphs in YAML, composing agent nodes (with LLM providers, memory, tools, thinking modes), human-in-the-loop nodes, Python executors, subgraphs, and control-flow edges with conditions and data transformations. The system supports dynamic parallel execution patterns (Map/Tree), multiple memory stores (simple, file-based FAISS, cloud Mem0), and extensible tooling via local Python functions or MCP servers. The architecture is a full-stack application: a FastAPI backend with REST and WebSocket endpoints for workflow execution, batch runs, artifact streaming, and session management; a Vue 3 frontend with a visual workflow editor (VueFlow), dynamic schema-driven forms, and real-time execution monitoring; a CLI entry point (`run.py`); and a Python SDK (`runtime/sdk.py`) for programmatic use. The codebase is well-structured with registry patterns for extensibility, declarative configuration dataclasses with auto-generated UI forms, and a modular runtime with pluggable node executors, edge conditions, and memory stores. While primarily an application, the project has strong research roots (OpenBMB lab, Tsinghua NLP) and includes academic references (ChatDev, Co-Learning, MacNet). It serves as both a production-grade tool for multi-agent automation and a research platform for experimenting with agent architectures, memory systems, and collaborative workflows. The repository includes 20+ example workflows demonstrating diverse use cases from software development and data visualization to 3D scene generation and deep research.
FlowPilot is a Chrome extension that automates the registration, verification, and delivery of accounts for AI platforms like OpenAI/ChatGPT, Grok (xAI), and Kiro (AWS). It handles email verification codes from a wide range of providers (Hotmail, iCloud, 2925, Gmail, etc.) and SMS verification through multiple services (5sim, HeroSMS, MaDao). The extension features a step-based workflow engine that composes registration flows dynamically based on signup method, payment mode, and delivery target. It includes an auto-run controller for batch account creation, account delivery to external APIs (CPA, SUB2API, etc.), and a contribution system for sharing accounts. Architecturally, it follows the standard Chrome extension pattern with background scripts, content scripts, and a sidepanel UI. The codebase is modular, using factory functions with dependency injection for testability. It has an extensive test suite with over 200 test files, though tests rely on a fragile function extraction method. The project is actively maintained with frequent commits and a versioned release cycle. The codebase demonstrates good engineering practices including a registry pattern for steps and mail rules, i18n support for Chinese and English, and reusable UI components. However, there are some maintenance concerns such as duplicate files, hardcoded external URLs, and dormant code paths for payment features.
Conductor is a production-grade workflow orchestration engine originally built at Netflix, now maintained by Orkes and the community. It provides durable execution for microservices, AI agents, and adaptive workflow graphs. The system defines workflows as versioned JSON DAGs with support for dynamic forks, loops, retries, timeouts, cancellation, pause/resume, and human-in-the-loop approvals. It has a modular Spring Boot backend with multiple persistence backends (PostgreSQL, MySQL, Cassandra, Redis, SQLite, Elasticsearch/OpenSearch) and event queue integrations (AMQP, NATS, SQS). The frontend is a modern React application with XState state machines, Material-UI, and a plugin system for extensibility. The codebase also includes a comprehensive AI integration layer with LLM providers (OpenAI, Anthropic, Gemini, etc.), vector databases, MCP tool calling, and the A2A agent-to-agent protocol. The architecture is microservices-ready with REST and gRPC APIs, a scheduler for cron-based workflows, and external payload storage (S3, GCS, Azure Blob). The project is actively maintained with a large community, but shows signs of technical drift due to multiple persistence backends, two coexisting UI codebases (legacy and modern), and some deprecated components.
MemoryBear is a production-oriented, multi-tenant AI agent platform whose core differentiator is a cognitive-inspired persistent memory system for LLM applications. The code — not the README — shows a full SaaS stack: a FastAPI backend (~60 domain controllers), a React/TypeScript admin console, an identity/auth service, an API gateway, a dedicated RAG knowledge service, and a Rust seccomp sandbox for code execution. The memory subsystem is the heart: pipelines extract statements, triplets, entities, emotions, preferences, and implicit habits from conversations into a Neo4j knowledge graph (projected to Elasticsearch through a transactional outbox), with ACT-R-inspired forgetting/reinforcement scheduling, LLM reflection/dedup, BERT scene-boundary detection, and hybrid retrieval at query time. This is an application, not a data-research repo, although the memory-science algorithms (forgetting curves, emotion classification, value ranking) are shipped as product features rather than research artifacts.
Direktiv is a workflow automation and serverless orchestration platform that enables users to define, execute, and manage workflows and services through a web UI or API. The system features a Go-based backend with React/TypeScript frontend, providing namespace-based multi-tenancy, event-driven triggers via CloudEvents and NATS, a plugin-extensible API gateway, and Kubernetes-native service deployment. Key capabilities include workflow execution (YAML/TypeScript), service management, secret/variable storage, file handling, Git-based namespace mirroring, and comprehensive monitoring with OpenTelemetry integration. The architecture follows a microservices-inspired monolith pattern with clear separation between API handling, data storage, execution engine, and gateway plugin systems.
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.
The Video Autopilot Kit is a Python-based framework for automating the creation of vertical shorts (YouTube Shorts, Reels, TikTok) and long-form videos. It provides three isomorphic production lines (teaching long-form, vertical shorts, interview show) each built as a knowledge layer → mechanical gate → one-command driver. The system centers around an Editkin v4 durable workflow contract that serves as the single source of truth for planning, evidence binding, audit, atomic apply, and render. Key features include an image-first material system requiring verified source media, a unified filter library of 38 reusable presets, and privacy-first design that avoids storing personal data in the repository. The architecture is modular, with specialized modules for asset management, color grading, audio processing, motion assets, art direction, scene auditing, and quality assurance. Entry points include CLI-driven pipelines for shorts (`src/shorts_autopilot.py`), interviews (`src/interview_autopilot.py`), long-form (`src/longform_maker/`), publishing (`src/publish_hub_cli.py`), and competitor teardown (`src/teardown.py`). The system relies on external tools like `ffmpeg`/`ffprobe` for video processing and optional OCR libraries for competitor analysis, with cross-platform compatibility handled via `src/platform_compat.py`. Despite its strengths, the kit faces maintenance challenges due to hardcoded assumptions about directory structures and platform specifications, as well as dependency drift risks from external tools like ffmpeg and OCR libraries. However, the inclusion of a release manager and self-updater helps mitigate some of these risks by providing deterministic builds and update mechanisms.
This repository is a production-grade, full-stack enterprise AI agent platform developed by iFlytek (Astron Agent). The system allows organizations to create, configure, deploy, and manage conversational AI agents (bots) with advanced workflow automation, knowledge retrieval (multi-strategy RAG), a plugin ecosystem (including MCP and HTTP tools), and multi-tenant team collaboration. The architecture is polyglot and service-oriented: a Java Spring Boot console backend handles bot lifecycle management, publishing to channels (API, WeChat, Feishu, marketplace), RBAC permissions, enterprise spaces, and notification systems. Python FastAPI core services power the workflow engine (20+ node types including LLM, code, decision, memory, knowledge), agent runtime with skill sandbox execution, knowledge indexing and retrieval, and plugin integrations (link server, RPA, MCP). A Go service manages tenant authentication and app key verification. The React + TypeScript frontend provides an interactive console for workflow design, model management, resource management (databases, knowledge bases, skill files), trace logs, and chat interfaces. External dependencies include multiple LLM providers (iFlytek Spark, OpenAI, Anthropic, Google, DeepSeek), MySQL, Redis, Kafka, S3-compatible storage, and Casdoor for SSO. The codebase demonstrates professional engineering with test coverage (JUnit, pytest, Jest), Flyway and Alembic database migrations, CI/CD pipelines, and a modular design using strategy, factory, and AOP patterns. However, it shows signs of complexity from polyglot maintenance, some duplicated frontend components, and legacy credential handling that has been addressed through targeted migration fixes. Overall, it is a well-maintained, actively developed platform suitable for mid-to-large enterprise deployments.
NovelForge is an AI-assisted long-form novel writing desktop application built with Electron, Vue 3, and FastAPI. It provides a card-based content management system where each story element (characters, scenes, chapters, items, concepts) is stored as typed cards with JSON Schema validation. The core differentiator is a custom workflow engine that uses a code-style DSL with marker comments (`#@node`) to define automation flows, supporting triggers, async execution, pause/resume, and streaming SSE progress. AI integration is deep: configurable LLM providers (OpenAI, Google, Anthropic, Qwen), structured generation via instruction streams (LLM outputs JSON instructions that the frontend executes incrementally), memory extraction from text into cards and a knowledge graph (dual backend: SQLModel or Neo4j), and a context assembly system with a rich template language. The frontend features a dynamic form renderer from JSON Schema, an inspiration assistant with tool-calling agents, a workflow agent that can autonomously edit workflow code via patches, and a chapter editor with AI continuation, polish, and review. The system is designed for professional novelists and supports full project lifecycles from tags and outlines to chapter drafts and quality reviews. The architecture is a typical two-tier application: the backend exposes RESTful APIs and SSE streams; the frontend is a single-page Vue 3 app bundled as an Electron desktop application (with web compatibility via mock). Notable engineering strengths include the workflow node registration pattern, the expression evaluator with safe eval, the memory extractor registry, and the dynamic form system—all of which are cleanly abstracted and potentially reusable across projects. The codebase is actively maintained, with regular commits, well-organized modules, and comprehensive bootstrap initialization for built-in prompts and card types. However, some features are experimental (workflow agent auto-patching), the custom DSL is format-sensitive, and the heavy reliance on LangChain may introduce version lock-in.
This is a Redmine plugin that allows administrators to define custom Ruby scripts that execute on lifecycle events (before_save, after_save, before_destroy, after_destroy, etc.) for various Redmine objects such as Issue, User, Group, Project, Attachment, TimeEntry, Version, WikiContent, and more. It uses Rails monkey-patching (prepend) to inject callbacks into Redmine models and controllers, and executes user-defined scripts via `instance_eval` in the context of the observed object. The plugin provides a web UI for managing workflows, supports import/export via XML, and includes a custom mailer for sending emails from workflows. It is well-structured, has comprehensive tests, localization support, and CI integration. The codebase is mature and actively maintained, targeting Redmine 6.x and Rails 8.1.
AgentTeams is a production-grade multi-agent orchestration platform that manages teams of LLM-powered worker agents on Kubernetes. It uses a central Manager controller that reconciles Custom Resource Definitions (Worker, Team, Human, Manager) with container backends (Docker, K8s), a Matrix chat layer for all agent communication, and an AI Gateway (Higress) for credential isolation. The system supports multiple worker runtimes (OpenClaw, QwenPaw, Hermes, DeepSeek Harness) through a plugin architecture, provides a CLI tool (`agt`) for resource management, and includes human-in-the-loop oversight via Matrix rooms. Key features include STS token-based credential management, role-based authorization, audit logging, file synchronization via MinIO/OSS, and a skill discovery and distribution system. The codebase is heavily tested with unit, integration, and end-to-end tests, but shows moderate technical drift due to legacy runtime coexistence (CoPaw EOL migration) and some duplicated shell scripts across runtimes.
Dograh is an open-source, self-hostable voice AI platform for building, deploying, and monitoring voice agents. It provides a visual workflow builder (node-based graph editor) for designing call flows, with deep integration across telephony providers (Twilio, Plivo, Vonage, Telnyx, Cloudonix, Vobiz, Asterisk ARI), STT/TTS/LLM models (OpenAI, Deepgram, Google, Azure, Cartesia, ElevenLabs, etc.), and real-time audio pipelines via the Pipecat library. The backend is a FastAPI monolith with PostgreSQL, Redis, and object storage, complemented by separate worker services (ARQ for background tasks) and a campaign orchestrator for bulk outbound calling. The system supports multi-tenancy, call recording, knowledge base/RAG, QA analysis, embeddable chat widgets, and an MCP server for AI-assisted workflow management. SDKs for Python and TypeScript enable programmatic workflow creation and outbound call triggering. While the core is a full-stack application, it includes evaluation frameworks for STT providers and post-call QA analysis, giving it a hybrid character but dominated by production application concerns.
SpiffWorkflow is a Python library for executing BPMN 2.0 and DMN 1.x workflows. It provides a complete workflow engine with parsing, execution, serialization, and migration capabilities. The library supports standard BPMN constructs (gateways, events, subprocesses, multi-instance loops, call activities) and DMN decision tables with UNIQUE and COLLECT hit policies. It includes a pluggable Python script engine for evaluating expressions and scripts within tasks, and a comprehensive serialization framework with version migration for workflow persistence. The architecture is modular, with separate packages for core workflow specs, BPMN specifics, DMN integration, Camunda-compatible extensions, and Spiff-specific customizations. This is not a standalone application but a library intended for embedding in larger systems; it is used as the backend for the SpiffWorkflow frontend tools.
Exosphere is a reliability runtime for AI agents, providing a state-based workflow orchestration platform. The system consists of three main components: a Python SDK (`exospherehost`) for defining nodes and execution graphs, a FastAPI state-manager backend that handles state transitions, retries, cron triggers, and secret encryption, and a Next.js dashboard for monitoring and managing workflows. The architecture follows a microservices-like pattern within a monorepo, with the SDK acting as a client library and the state-manager as the central orchestration engine. Key capabilities include dynamic graph execution, infinite parallel fanout/unite patterns, configurable retry policies (exponential, linear, fixed with jitter), signal-based state pruning and re-queuing, and a graph-level key-value store. The system relies on MongoDB for persistence and uses Beanie ODM. The codebase demonstrates solid engineering practices with well-structured controllers, middleware for request ID propagation and exception handling, and a comprehensive test suite, though some tests are incomplete or skipped. The project is in early beta (v0.0.3b1) with active development and monthly releases.
O2OA is an open-source low-code development platform for enterprise collaborative office (OA) solutions. It features a JavaEE-based distributed architecture with a process engine, form/page customization, and business data services. The platform provides cross-platform support (Windows, Linux, Unix, ARM, etc.) and recommends production-ready databases like MySQL/Oracle/SQLServer. The system is organized into modular components including attendance, CMS, portal, process management, organization management, file handling, messaging, and AI integration. Backend uses Java with JAX-RS REST services, while frontend employs a custom MWF (O2OA Web Framework) alongside CKEditor for rich text editing. The architecture follows a layered approach with client-side UI components and server-side RESTful services, utilizing JPA for persistence and DTO patterns for data transfer.
Solace Agent Mesh is a production-grade, full-stack platform for building, deploying, and managing event-driven multi-agent AI systems. It comprises a Python backend (FastAPI, Google ADK, SAC), a React/TypeScript web UI (chat, projects, workflows, prompts, scheduled tasks), a Click-based CLI, and a Docusaurus documentation site. The system uses a microservices-like architecture with separate gateway (HTTP/SSE), platform service (model configuration, storage), and agent components, all communicating via the Solace event broker and the A2A (Agent-to-Agent) protocol. Key entry points include the CLI (`solace-agent-mesh`), the web UI (`/chat`, `/projects`, `/workflows`), and the REST API (`/api/v1`). Data flows involve user requests via SSE, agent orchestration with workflow DAGs, tool execution (built-in, MCP, OpenAPI), artifact storage (S3/GCS/Azure/filesystem), and session persistence in SQL databases (SQLite/PostgreSQL/MySQL). External dependencies include a Solace broker, LLM endpoints (LiteLLM), optional databases, and cloud storage services. The repository is marked as deprecated (Development Status 7 – Inactive), with the new version being a desktop application.
OptimaJet.Workflow is a .NET workflow engine library designed for business process management and automation. It provides a core runtime for defining, executing, and persisting workflow schemes (process definitions) across multiple database backends including SQL Server, MySQL, PostgreSQL, Oracle, SQLite, and MongoDB. The engine supports multi-tenancy, version control of workflow schemes, and includes a visual HTML5 designer for creating workflows. It is architected as a library/framework layer, intended to be integrated into applications rather than running as a standalone service. The codebase demonstrates a hybrid architecture, combining modern .NET Core samples with legacy .NET Framework implementations using Windows Workflow Foundation (WWF). This duality is evident in the presence of both `OptimaJet.Workflow.Core` (modern) and `Other/WorkflowFoundation` (legacy) directories. The library relies heavily on the Provider Pattern for persistence, allowing developers to swap database backends via dependency injection. It also features a robust migration framework (FluentMigrator for SQL, custom scripts for MongoDB) to manage schema evolution, particularly around multi-tenancy support (`TenantId` columns). While the library offers solid engineering patterns like generic database abstraction (`DbObject<T>`) and async/await persistence, the codebase suffers from significant technical drift. It includes deprecated technologies such as `System.Workflow` and WCF, and sample applications rely on older versions of frameworks like ABP (v6.5.0) and .NET Core 2.2. The duplication of persistence logic across five different database providers increases maintenance overhead and the risk of schema drift between implementations.
Vizra ADK is a retired Laravel package (final release v0.0.48) that provides a comprehensive framework for building AI agents with LLM orchestration, tool systems, memory, workflows, evaluations, tracing, and MCP integration. It is not a standalone application but a library designed to be installed into Laravel projects. The architecture follows Laravel package conventions: service providers, facades (Agent, Workflow), Eloquent models, Artisan commands, Livewire components, and database migrations. Key entry points include the Agent facade for running agents, the Workflow facade for orchestrating multi-agent workflows, and a set of CLI generators for scaffolding agents, tools, and evaluations. The system supports multiple LLM providers via Prism PHP, persistent memory (session, user, vector), tool chaining, sub-agent delegation, and an evaluation framework with LLM-as-a-Judge assertions. The codebase is well-structured with strong test coverage (Pest + Mockery), but the package is officially retired with no further development, leading to high technical drift as dependencies age out. Despite retirement, many components (e.g., ToolChain, VersionablePrompts, AgentVectorProxy, HasLogging) are self-contained and demonstrate reusable patterns suitable for extraction into new projects.
Agentspan is a comprehensive, multi-language agent orchestration platform built on top of Orkes Conductor. It provides SDKs in Python, TypeScript, Java, and C# for defining, running, and deploying AI agents with rich features: tools, guardrails, multi-agent strategies (handoff, sequential, parallel, router, swarm, plan-execute), streaming, human-in-the-loop, code execution in sandboxes, credential management, and scheduling. The architecture follows a server/client pattern where agent definitions compile to Conductor workflows, with a Java/Spring Boot server, a Go CLI with a Bubble Tea TUI, and a React/Material-UI web interface for visual workflow editing and execution monitoring. The repository is archived (read-only) as of August 2026, with its active development having merged into the Orkes Conductor repository. Despite being archived, the codebase represents a mature, well-crafted multi-agent runtime with extensive examples, comprehensive testing, and clean abstractions across all supported languages. The SDKs demonstrate consistent design patterns—agent composition via strategies, tool registration via decorators/annotations, guardrail chains with composable termination conditions, and thread-safe credential injection—making the codebase a valuable reference for building agent orchestration systems.