This article explores Prism, a plugin-based red team agent built on DeepSeek Harness and Cordis, detailing its architecture, safety guardrails, shared state management, tool integration (Nmap, Nuclei, Burp Suite, Metasploit), and dashboard for autonomous asset discovery and vulnerability verification within defined security boundaries.
This article explains DeepSeek Harness's two core extension mechanisms: services for reusable logic shared across plugins, and event listeners (ctx.on) for observing, intercepting, and rewriting execution flows, with complete code examples for a summarization service and a guardrail plugin.
The article explains why process orchestration is essential for backend systems serving multiple business lines, demonstrating a flow engine with process, success, rollback, and callback nodes that eliminates if-else chains, ensures code isolation, and enables safe extension, with code examples and a Git repository reference.
Chain of ResponsibilityMulti-tenancyPlugin Architecture
A comprehensive 14k-word technical deep-dive into MyBatis covering its semi-automatic ORM design, four core components (Executor, StatementHandler, ParameterHandler, ResultSetHandler), dynamic SQL parsing via SqlNode trees, two-level caching, plugin mechanism, Spring integration, and source-code-level analysis of 11 key classes with practical best practices and common pitfalls.
This article dissects Claude Code's Mods system — functional hooks that let developers intercept tool calls, prompts, and UI rendering — explaining the event chain, four built-in mods (diff, agents-md, sec-default, telemetry), security boundaries, and when to choose mods over Skills or MCP.
This guide walks through building a Claude Code Mod from scratch using the Token Weather example, explaining the plugin structure, event hooks, state management, UI rendering, and the three hook patterns (observe, rewrite, answer), while comparing Claude Code's host-centric runtime control plane with DeepSeek Harness's plugin-first architecture.
Open File Viewer v0.1.48 now supports legacy Office binary formats (doc, xls, ppt) alongside modern formats, offering a framework-agnostic file preview solution for Vue, React, Svelte, and Vanilla JS with 274 supported extensions and a plugin architecture that handles complex rendering fallbacks.
This article reviews three cutting-edge AI open-source projects: DeepSeek's plugin-based Agent runtime framework (deepseek-harness), archify for generating verifiable architecture diagrams directly in coding agents, and colibri, a pure C inference engine enabling 744B MoE models to run on 25GB RAM via disk streaming.
This article dissects DeepSeek Harness, an MIT-licensed agent framework where everything is a plugin, explaining its Cordis-based spatiotemporal composability, internal execution loop, and trade-offs against Claude Code's integrated approach.
The article argues that a composable agent runtime alone is insufficient for production; it requires action contracts for external side effects, version binding with impact analysis, and independent audit evidence for task completion, with clear responsibility boundaries across runtime, plugins, action gateways, and human operators.
Anthropic rolls out Claude Mods, a plugin system built on Function Hooks that lets developers intercept tool calls, modify UI components, and extend Claude Code like middleware, with community demos including Tetris and security tests revealing both power and early reliability gaps, signaling a shift in coding agent competition from model strength to platform extensibility.
The article reviews the developer preview of DeepSeek Harness's official desktop client, highlighting its plugin-based architecture powered by Cordis, deep integration with DeepSeek V4.1 Flash model achieving 99% cache hit rate, and comparative performance against other harness integrations like Pi Agent.
This article details DeepSeek Harness's plugin-centric architecture (Agent = Model + Harness) and reviews 10 key plugins — including dsh-market, modlens, dsh-web-ui, and OpenViking — with installation commands, repository links, and practical use cases for vision, UI, search, memory, cost tracking, and document analysis.
This article details DeepSeek Harness's general settings including permission policies, model adapters, and plugin architecture, then compares four agent presets—Standard, Minimal, PTC, and Creative—showing their token usage, toolsets, and ideal use cases with a 3D game benchmark.
AI agent configurationCreative ModeDeepSeek Harness
Huawei Cloud DWS introduces an AI diagnostic feature using a three-layer Agent-CLI-Skill architecture that automates the data warehouse troubleshooting loop—collection, analysis, root-cause reasoning, and reporting—turning expert DBA knowledge into reusable, pluggable skills that cut diagnosis time from hours to minutes.
This tutorial introduces DeepSeek Harness, a plugin-based AI agent framework, covering its design philosophy, quick and source-code installation methods using Node.js and pnpm, and demonstrates building a 3D racing game in one shot compared to multi-iteration alternatives.
A deep technical analysis of DeepSeek Harness (dsh), an open-source agent runtime that acts as a model exoskeleton — plugin-based architecture, 1,546-line core loop, session-log-driven design, and built-in data loop via HTTP extensions.
AI agent runtimeDeepSeek HarnessPlugin Architecture
The article compares three open‑source agent harness frameworks—DSH, OpenCode, and Pi—detailing their architectural philosophies, customization mechanisms, security models, maturity levels, and recommending which to choose based on a developer's relationship with the agent.
DeepSeek Harness, released as an open‑source developer preview, applies category‑theoretic concepts such as objects, morphisms, functors, and monads to turn every plugin into a composable arrow, giving AI agents a mathematically guaranteed, predictable execution framework.
The article walks through installing DeepSeek Harness, configuring its plugin‑based architecture, and using it to automatically generate a full‑featured Vue‑based Markdown editor, demonstrating the framework’s ease of setup, detailed execution trace, and iterative refinement capabilities.
This article compares three open-source AI coding agent harnesses—Pi, DeepSeek Harness, and Codex Harness—analyzing their architectures, extension mechanisms, and enterprise selection criteria to help teams build customized AI coding platforms.
The article analyses Pi’s Harness v2 redesign, explaining how persistent transaction‑style recording, tool‑result pruning with spill, replay policies, and separated storage of conversation, runtime state, and usage enable an agent to survive process crashes and still prove which steps were completed.
The article provides a hands‑on review of DeepSeek Harness, detailing its plugin‑centric design, installation steps, UI features, permission and run modes, and extensive performance comparisons against Codex + GPT‑5.6 Sol across several animation and game tasks, followed by a cost analysis and final usability insights.
The DeepSeek Harness preview, released in August 2026, quickly amassed over 150,000 GitHub stars as developers embraced its plugin‑centric agent runtime that separates the model (brain) from the harness (hands), offering a flexible, auditable framework that outpaces traditional AI coding assistants.
DeepSeek Harness, the fast‑growing AI Agent framework with over 129 K GitHub stars, now offers two community‑built desktop clients—deepseek‑harness‑desktop and dsh‑desktop—each with distinct features, installation paths, plugin ecosystems, and benchmark performance, helping developers choose the best setup for their needs.
DSH Desktop is an open‑source Thin‑Electron client that treats desktop capabilities—window, tray, terminal, updates, and configuration—as first‑class DSH plugins via the Cordis host, offering installation rollback, startup recovery, and safeguards against accidental plugin disabling.
While most AI‑enhanced IDEs still extend VS Code’s editor‑centric model, DeepSeek Harness starts from an Agent runtime, treating everything as a plugin and positioning the development tool’s core around an Agent Host, raising questions about whether the editor will remain the central hub.
DeepSeek Harness, released on August 13 under the MIT license, is an open‑source, plugin‑centric agent runtime that offers four operation modes, builds on the Cordis/Koshi framework, provides full model‑agnostic support, and includes detailed logging for reproducible AI workflows, while noting current limitations.
This step‑by‑step guide shows how to set up DeepSeek Harness, configure the API key and workspace, run a Java project analysis, choose the appropriate agent mode, and decide when to install from source, providing practical tips for Java developers.
DeepSeek Harness (DSH) introduces a plugin‑centric AI Agent architecture that treats the Agent as the emergent result of configurable plugins, leverages the Cordis spatiotemporal composability framework, offers four distinct runtime modes, and is compared against Claude Code, Codex, LangGraph and AutoGen, highlighting its strengths, practical usage, and current limitations.
The author launches DeepSeek Harness locally, issues a single command to create a Markdown‑enabled demand‑management menu, examines the resulting UI and token consumption, and explains DSH's dual nature as a runnable coding agent and a plugin‑centric development framework.
The article analyzes how DeepSeek Harness adopts an agent‑centric, all‑plugin architecture that could shift the developer‑tool ecosystem’s focus from the traditional VS Code editor model to a more flexible runtime, comparing it with Cursor’s VS Code‑fork approach and outlining current limitations and future possibilities.
The article analyses the complexity of managing Harness assets (rules, skills, hooks, configs) across multiple product lines, agents and devices in commercial clients, and details a four‑stage evolution—from single‑file sync to full ACX repository pluginization—providing a reproducible, unified governance model with daily auto‑updates and clear host integration.
This article examines how the 2,000‑line Cordis micro‑kernel underpins DeepSeek Harness’s “everything is a plugin” architecture, detailing its five plugin primitives, six architectural practices, vendor‑embedding strategy, typed events, service injection, effects, and runtime self‑modification, and evaluates the resulting modularity, replaceability, hot‑plugability, extensibility, testability, evolvability, and auditability.
DeepSeek Harness v0.1, an MIT‑licensed framework driven by Cordis, treats models, tools, skills and even the execution loop as interchangeable plugins, flattening previous layered architectures, enabling true self‑evolution of agents while exposing new risks and open questions about safe modification and evaluation.
DeepSeek Harness, an open‑source agent framework built on the Cordis meta‑framework, treats models, tools, skills, sessions, sandboxes, loops and UI as interchangeable plugins, enabling dynamic composition, fine‑grained token efficiency and full chain‑of‑thought tracing while avoiding the lock‑in typical of other AI model frameworks.
The author evaluates the newly open‑sourced DeepSeek Harness by running its web, headless, Python SDK and ACP interfaces, comparing its plugin‑based agent runtime, trajectory logging, and token usage against Kimi Code on identical tasks, and draws practical conclusions for developers and everyday users.
The article analyzes DeepSeek Harness's plugin‑based architecture, its Cordis core for reversible side‑effects, the four operational modes—including a creation mode that lets agents dynamically add or remove components—and the underlying research paper that formalizes spatiotemporal composability for self‑modifying AI agents.
This article dissects DeepSeek Harness (DSH), showing how its plugin‑based runtime graph and event‑stream architecture let agents dynamically compose capabilities, manage lifecycles, and recover from interruptions, while comparing its design to the Pi coding agent and outlining practical presets and configuration steps.
DeepSeek Harness introduces a fine‑grained, fully plug‑in‑based architecture that makes LLM agents transparent, traceable and highly customizable, allowing developers to replace or extend any component—from model adapters to the main loop—thereby exposing optimization opportunities that were hidden behind traditional monolithic frameworks like LangChain.
On August 13, DeepSeek simultaneously released the flagship V4‑Pro model and open‑sourced its Harness runtime, illustrating the “Agent = Model + Harness” paradigm; the article details the model’s pricing, performance, new features, the plugin‑centric design of Harness, usage steps, community ecosystem, and broader AI‑agent implications.
On a single day, Google, DeepSeek, and MiniMax unveiled Gemini 3.7 Flash, V4‑Pro, the Harness runtime, and Music 3, each illustrating how AI is shifting from headline‑grabbing benchmarks toward cost‑effective agents, plug‑in runtimes, and controllable content generation for real‑world workflows.
DeepSeek Harness (DSH) is an open‑source, MIT‑licensed agent framework that treats every capability as a plugin, provides an append‑only execution log for full traceability, offers multiple built‑in modes including headless operation, and distinguishes itself from Claude Code/Codex by delivering a truly modular, model‑agnostic execution layer.
The article provides a detailed technical walkthrough of DeepSeek Harness (dsh), highlighting its plugin‑centric architecture, full‑trajectory visibility, performance metrics, installation steps, four UI modes, the underlying Cordis framework, the capability‑seam design, and community‑contributed plugins, all illustrated with concrete examples and code.
The article dissects DeepSeek Harness, an open‑source AI agent framework, showing how it solves the long‑standing “dynamic composition” problem by introducing revertible effects and reactive coeffects, enabling safe self‑modification without downtime or hidden side‑effects.
DeepSeek has released its Harness framework as open source, a TypeScript‑heavy (97.1% of code) Node.js project that provides a plugin‑based agent runtime, integrates multiple model providers, and is currently in Developer Preview, offering a one‑command start but warning about upcoming breaking changes.
DeepSeek Harness, now in a globally open developer preview under the MIT license, introduces a thin Cordis core and a fully plugin‑based architecture with over 130 interchangeable capabilities, four preset modes, exhaustive session tracing, and simple npx or source‑code startup, marking a shift from model‑only competition to agent‑infrastructure innovation.
The DeepSeek Harness SDK, now open‑source, offers a completely pluggable architecture for building AI agents, provides four preset modes, multiple entry points, a fail‑closed security model, and is underpinned by a rigorous academic paper on spatiotemporal composability that formalizes reversible effects and reactive coeffects.
DeepSeek has released the V4 Pro model and, hours later, open‑sourced the DeepSeek Harness runtime, a plugin‑based agent framework that connects models to files, terminals, tools, and workflows, offering extensible architecture, risk controls, and a Python SDK while still in developer preview.
The article examines the challenges of fragmented front‑end file preview solutions and introduces Open File Viewer, a unified SDK that supports 110 file formats, offers a customizable toolbar, and provides a plugin architecture for extensible, framework‑agnostic previews in modern web applications.
The article explains why rebuilding and redeploying a whole JAR for small rule or script changes is inefficient, distinguishes development‑time hot reload from production hot deployment, and shows how a plugin architecture using URLClassLoader and ServiceLoader provides granular, rollback‑friendly, and extensible updates while warning of classloader leaks and dependency conflicts.
The article explains how the pi agent’s extension system resolves the tension between core stability and capability extensibility by using inversion of control, dependency injection, adapter and event‑driven patterns, two‑phase initialization, and concrete Python implementations, while comparing it with other plugin architectures.
The Claude Code plugin ecosystem has exploded to over 9,000 extensions with the leading plugins surpassing 50‑100k installs; this guide explains the five plugin architecture types, evaluates six functional directions, and recommends tailored plugin combos for full‑stack engineers, frontend developers, solo creators, and team leads.
The article explains how excessive if‑else branching in a middle‑platform backend leads to maintenance nightmares, and demonstrates how a workflow engine combined with a plugin extension mechanism provides clean code isolation, flexible business extensions, and reliable execution flow.
DeepTutor, an open‑source AI tutoring platform from HKU’s Data Science Lab, introduces an agent‑native architecture with a two‑layer plugin model, persistent memory, and multimodal capabilities, offering developers and learners a customizable, interactive educational experience via CLI and web interfaces.
Google’s new Java Agent Development Kit 1.0 provides a structured, plugin‑based framework that lets Java backend teams embed large‑language‑model agents, manage context and token limits, integrate secure tools, persist state, and enable cross‑language Agent2Agent collaboration without rewriting existing architectures.
This article explains how Spring Boot’s plugin mechanisms—such as Java SPI, custom configuration files, and Spring Factories—enable module decoupling, extensibility, and easy third‑party integration, and walks through complete code examples and a real‑world SMS‑sending case study.
This article walks through the inner workings of the open‑source tRPC‑Go RPC framework, detailing the client invoke lifecycle, selector reporting, codec serialization, transport handling, server processing, plugin architecture, streaming support, and performance‑oriented design choices for Go backend developers.
In eight days the author used a recursively evolving AI agent orchestrator to generate 40,000 lines of TypeScript, 17 plugins, and 3,288 tests, demonstrating parallel task decomposition, self‑healing CI, and a fully pluggable architecture that automates code review and deployment.
This article explains the design of OpenClaw’s Gateway – the control‑plane that unifies channel access, message routing, agent provisioning, distributed node execution, session isolation and hot‑reload configuration – and shows how each piece enables a scalable, governed AI‑agent platform.
The open-source openclaw-china plugin (2.2K Stars) adds seamless AI assistant integration for DingTalk, QQ, and Enterprise WeChat, offering multi‑format messaging, proactive pushes, streaming responses, and a layered plugin architecture that keeps the core OpenClaw framework untouched.
This guide explains Java's Service Provider Interface (SPI) mechanism, compares it with APIs, and walks through creating a multi‑module Maven project that defines SPI interfaces, implements plugins, loads external JARs with a custom class loader, and integrates the plugins into a Spring Boot application for dynamic authentication.
This article explains Java's Service Provider Interface (SPI), compares it with traditional APIs, and provides a step‑by‑step guide to create a multi‑module Maven project, custom class loader, and Spring Boot application that dynamically loads authentication plugins from external JARs.
This article walks through refactoring a tightly‑coupled Go API gateway into a lightweight, plugin‑driven microkernel architecture, detailing the motivation, design of core abstractions, code examples, and the resulting gains in extensibility, maintainability, and flexibility.
This article explains how to implement plugin mechanisms in Java using SPI and Spring Boot's spring.factories, covering benefits, common approaches, detailed code examples, custom configuration loading, and a complete real‑world case study to guide developers in building extensible applications.
This article explains why over‑designed code hampers maintainability, introduces flow engines and plugin extension engines as solutions for isolating business logic, and provides practical guidance on configuring flow chains, defining nodes, and executing processes with Java examples.
This article details the technical design of the Apex VSCode plugin, covering its background, overall architecture, activation flow, SSO authentication, rule‑knowledge‑base engineering, remote Webview integration, version orchestration, project service handling, logging, and future enhancements for AI‑driven development efficiency.
The author shares how they built a low-code platform in two months with 90% AI-generated code by adopting an architect-and-tamer role, decomposing tasks, feeding context, and iterating with tests across four core modules.
This article introduces ZHPluginSDK, a powerful Swift‑based plugin framework that supports type‑safe registration, flexible message dispatch, multi‑container composition, synchronous messaging, and environment variables, providing a comprehensive guide with architecture diagrams, feature lists, and extensive code examples for mobile developers.
This article explores the challenges of traditional monitoring in cross‑platform front‑end development and presents Hippy Monitor’s plugin‑based architecture, intelligent sampling, multi‑environment configuration, and AI‑driven roadmap as a comprehensive solution for full‑chain observability and proactive performance management.
The article explains how excessive if‑else branching in multi‑business systems harms maintainability, argues for using a flow engine and plugin extension engine to achieve code isolation and extensibility, and demonstrates the approach with concrete Java examples from the open‑source MemberClub project.
This article explains the concept and benefits of plugin-based development in Java, introduces common implementation approaches such as SPI, custom configuration, and Spring Boot's spring.factories, and provides step‑by‑step code examples and a real‑world case study to help developers build extensible backend systems.
This article explains how to implement a plug‑and‑play development model in Spring Boot using Java SPI, custom configuration, and Spring Factories, providing step‑by‑step code examples, a multi‑module microservice case study, and practical guidance for creating extensible, decoupled applications.
Concept Plugin 2 introduces a lightweight, annotation‑driven plug‑in system for Spring Boot that separates core business logic from device integration code, enabling independent testing, dynamic loading, and easy configuration through a management UI, while supporting various injection patterns, nested dependencies, and Spring features.
This article explains how to design and implement a plugin mechanism in Java and Spring Boot, covering the benefits of modular decoupling, common implementation patterns such as ServiceLoader and custom configuration files, and practical code examples for building extensible backend services.
This article explains Java's Service Provider Interface (SPI) mechanism, compares SPI with traditional APIs, and provides a step‑by‑step guide—including Maven project setup, interface definition, implementation classes, custom class loader, and Spring Boot integration—to build a dynamic plugin system for backend applications.
The article explains how the MemberClub open‑source project leverages a flow engine and plugin extension engine to achieve business code isolation and extensibility, illustrating the problems of if‑else branching in multi‑business systems, providing configuration examples, execution principles, and Java code snippets.
The article explains how to avoid tangled if‑else code in a Java backend by using a process engine and a plugin‑based extension point framework, detailing interface definitions, annotations, loading mechanisms, and usage examples drawn from the open‑source MemberClub project.
The article explains why simple code quickly becomes unmanageable in multi‑business systems, introduces workflow engines and plugin extensions as solutions, and demonstrates configuration, node definition, execution, and the underlying Java implementation to achieve clean, isolated, and extensible business logic.
This article explains why plugin mechanisms improve modularity, extensibility, and third‑party integration, then walks through practical Java SPI implementations, custom configuration loading, and Spring Boot’s spring.factories approach, providing complete code examples and step‑by‑step guidance for building a robust plugin system.
This article explains Java and Spring Boot plugin mechanisms, covering benefits, common implementation strategies such as ServiceLoader, custom configuration files, and Spring Factories, and provides practical code examples and a real‑world case study to illustrate extensible architecture design.
iLogtail, a lightweight log collector, has transformed over 13 years from a simple log‑gathering tool into a full‑stack, cloud‑native observability platform, introducing Go plugins, high‑performance C++ pipelines, SPL processing, modular architecture, and advanced self‑monitoring, reflecting broader trends in data collection technology.
This article chronicles iLogtail's 13‑year journey—from its 2013 inception as a basic log collector to a fully open‑source, cloud‑native observability platform—highlighting technical milestones, emerging trends in log agents, architectural innovations, performance breakthroughs, and future directions.
This article describes how to create a feature‑rich, extensible React admin panel by implementing a convention‑based routing system, unified navigation with menu and draggable tabs, a Redux‑centric state management strategy, and a modular plugin architecture, while also integrating MUI, Next.js, and AI‑assisted code generation.
This article explains the concept of plugin‑based development, outlines its benefits such as modular decoupling and extensibility, and provides detailed Java implementations using ServiceLoader, custom configuration loading, and Spring Boot’s spring.factories mechanism with complete code examples and practical deployment steps.
Lexical is a highly extensible, framework‑agnostic JavaScript text‑editing library created by Facebook, offering a lightweight core, plugin architecture, custom nodes, themes, and state management for building rich‑text, collaborative, and native‑like editors across web applications.
Mako is a lightning‑fast, zero‑configuration web bundler whose Rust core and parallel Node.js pool handle loading, parsing, transforming, dependency analysis, and asset generation across JavaScript/TypeScript, CSS, assets, and more, while offering HMR, code‑splitting, plugin extensibility, and production‑grade reliability.
This article explores VS Code’s plugin ecosystem, detailing its capabilities, multi‑process architecture, Electron foundation, extension loading mechanisms, and step‑by‑step guidance for building, debugging, and publishing extensions—including language server integration—while highlighting best practices for efficient development.
AutoDev for VSCode extends the AI-driven coding assistant from IntelliJ IDEA to the VSCode ecosystem, detailing its unified architecture, dependency-inverted plugin system, event-driven UI, core interfaces, test-generation prompts, and the 0.1.0 release features such as custom AI commands, test and comment generation.
This article explains why plugin architectures boost modularity, extensibility and third‑party integration, then walks through multiple Java‑based implementation techniques—including ServiceLoader, custom configuration files, and Spring Boot’s spring.factories—showing complete code samples, runtime screenshots and a real‑world microservice case study.
Apache Shenyu is a high‑performance, open‑source API gateway built on the reactive WebFlux framework that unifies API management and security through a flexible plugin architecture offering authentication, traffic control, circuit breaking, request rewriting, logging, and monitoring, while delivering extensibility, low latency, and future cloud‑native integration.
This document details the motivation, technology selection, overall architecture, data‑plane and control‑plane design, key implementation challenges such as HTTPS interception, precise URL monitoring, and plugin extensibility, and outlines future directions for scaling service mesh within the organization.
Facing low development efficiency, poor CPU utilization, and fragile fault tolerance, the QQ Browser content ingestion team rebuilt a 93‑service microservice system into a single‑process, plugin‑driven architecture, achieving up to 13‑fold throughput gains, 10‑fold batch‑processing speedups, and dramatically reduced lead times and code complexity.
This article describes how an enterprise leveraged GPT‑3.5 and other large language models to create a private knowledge base, design prompt engineering, implement plugin extensions, and build a secure, scalable backend and front‑end integration platform that enables AI‑driven customer‑service assistants across multiple business lines.
This article details the complete redesign of QQ Browser's content ingestion system, explaining why the original micro‑service architecture caused low efficiency and performance, and how a zero‑base redesign using a monolithic service, plugin framework, fault‑tolerant pipelines, and thread separation dramatically improved throughput, latency, and developer productivity.
tRPC is an open‑source, plugin‑based RPC framework that separates core communication, service‑governance, and invocation layers, supports extensible codecs, naming, config, metrics, logging and tracing plugins, offers an Admin API, and demonstrates high throughput and low tail latency in cloud‑native deployments.