14 articles
Explore AI agent frameworks, tool use and task automation with framework comparisons, setup guides and practical examples.
Distinguish chat assistants, fixed workflows and agents that choose and use tools. Select a framework for the task, then review permissions, runtime requirements and human checkpoints before testing multi-step work.
Manus has introduced Cue alongside Manus 2.0, a personal AI Agent with its own phone number, email address, wallet, and computer. The product aims to move AI from one-off answers toward persistent, real-world assistance.
OpenClaw 2.0 (v2026.8.1) rebuilds the web console, adds cross-device and cloud task continuation, stronger memory, session branches, approval controls, and agent collaboration.
DeepSeek Harness is an open Agent runtime from DeepSeek AI, powered by Cordis. It turns models, tools, sessions, sandboxes, and the Agent Loop into composable plugins.
DeepSeek-V4-Flash reaches 82.7 on Terminal-Bench 2.1, beats GLM-5.2, approaches Claude Opus 4.8, and costs just ¥1 input / ¥2 output per million tokens for agents.
Anthropic has released Claude Opus 5 for complex agentic coding and enterprise work, with 1M context, 128K output, and API pricing at half the cost of Claude Fable 5 today.
Microsoft Phi-Ground-Any is a 4B GUI grounding model for computer-use agents. It outputs click points on a 1680x1008 canvas to help desktop and web agents target UI elements more reliably.
DeepSeek-V4 released with 1M context as standard! Agent capabilities surpass open-source models, reasoning performance rivals GPT-4o and Claude Opus.
OpenClaw supply chain attack: 1184 malicious skills, 135K vulnerable instances. Researcher breaks ecosystem in 8hrs. Full analysis of AI Agent era's npm poisoning 2.0 with survival guide.
DeepSeek-V3.2 is the latest open-source LLM rivaling GPT-5 and Gemini-3.0-Pro, featuring DSA sparse attention and gold-medal results in IMO/IOI 2025.
DeepSeek V3.1 adds Think/Non‑Think hybrid reasoning, stronger coding and Agent abilities, 128K context and MoE efficiency, plus Anthropic API support for Claude Code.
Build AI agents without code using LangChain Open Agent Platform. Features RAG, tool integration & multi-agent collaboration. Open-source, developer-friendly, and enterprise-ready.
Explore four pillars of modern AI: Agent, RAG, Function Call, and MCP. Learn how these technologies work together to enhance AI capabilities through practical examples and analogies.
Manus vs OpenManus: A clash between commercial and open-source AI Agents. While Manus's invitation codes sell for $7,000, MetaGPT's 3-hour open-source recreation sparks debate on tech innovation.
CogAgent-9B: A 9B-parameter GUI agent by Zhipu AI and Tsinghua University that excels in interface understanding and automation, outperforming other models in MM-Vet and more benchmarks