agent architecture
40 articles · 15 co-occurring · 1 contradictions · 5 briefs
LLM → agent → agent harness → coding harness" — Article explicitly describes the architectural layers of a coding agent, showing how components stack from LLM through harnesses.
[INFERRED] "An agent wouldn't do this. An agent would relentlessly pursue its task until it's done correctly. You know what would do this? A fancy autocomplete." — Article challenges the framing of LLMs as agents, arguing they behave more like autocomplete systems that lack persistent task pursuit and correction mechanisms.
MCP is the infrastructure layer that enables autonomous agents to act on the external world. Without it, agents are confined to reasoning-only mode.
an agent as a system using an LLM to decide the control flow of an application" — Article directly defines and structures LangChain agents as LLM-driven control flow systems, core architecture pattern
four key components are required: an agent core, a memory module, agent tools, and a planning module" — Article explicitly defines the foundational architectural components of LLM agent systems
LLM → agent → agent harness → coding harness" — Article explicitly describes the architectural layers of a coding agent, showing how components stack from LLM through harnesses.
Isolating context - splitting it up to help an agent perform a task." — Introduces context isolation as an architectural pattern for agent design, adding a new dimension to how agents structure their
先让 Agent 完成几项真实任务（真实场景比合成测试更能暴露问题）。在LangFuse中打开对应trace，重点关注："where it gets confused"" — Article provides concrete debugging methodology: running real tasks and analyzing where agent reasoning fails, d
Build complete ReAct and MDP agent loops with real tool execution" — Covers both ReAct (Reasoning-Acting) and Markov Decision Process agent loops with practical tool execution implementation
Introducing our new research direction on Recursive Language Models. We are sharing our initial experiments showing the promise of RLMs." — RLMs represent a specific architectural innovation for agent
agentic support is new baseline for any SaaS" — Article asserts that agentic capabilities have become essential architectural requirement for SaaS platforms
Article explicitly contrasts Claude Code's agent-system design (better context chaining) against IDE extensions and CLI tools, showing agent architecture as superior for context preservation.
A decade of information retrieval and three years of agent harness engineering, poured into a local-first knowledge system" — napkin represents evolved thinking on agent harness engineering with novel
Agentic AI creates goal-driven autonomous agents that can observe their environment, plan actions, execute tasks, and maintain long-term memory. Unlike standard LLMs, which only respond to prompts, ag
A complete enterprise agent stack in 2026 will use all four protocols: MCP for tool access, A2A for agent coordination, and ACP or UCP for commerce transactions depending on the environment. The four
build production-ready AI agents" — Article demonstrates practical implementation patterns from Anthropic and LangChain for production agent systems
years of experience in building AI systems is that Contexts are underrated. It is clearly affecting the way your RAG or AI Agents perform." — Article provides evidence-based insight that agent perform
Build your first agent with models, tools, memory, and streaming" — Tutorial explicitly demonstrates building agents as a core learning objective, showing models, tools, and memory integration.
Using the Bun runtime—known for its extreme speed in executing JavaScript—Claude Code monitors specific plugins (currently Telegram and Discord)." — Claude Code Channels demonstrates a concrete agent
Before MCP, every AI agent needed unique, hand-coded integration logic for every external service it accessed. This created an unsustainable development burden and vendor lock-in. MCP solves this by e
Tool is built for observing agent systems with tool calling, thinking, and delegated background jobs. Reveals that production agents need observability infrastructure as a first-class concern.
today agents are good at "OLTP" tasks RLMs make them good at "OLAP" tasks as well" — Article extends understanding of agent capabilities by introducing OLAP vs OLTP distinction, showing how RLMs expan
Prescribes specific minimal viable agent pattern based on message handling and JSON parsing
Article focuses on context engineering specifically for agents, suggesting agent design patterns are the application domain
Demonstrates practical agent architecture pattern (Model + Harness) with specific application to coding agents, showing how to structure trust-building systems.
[INFERRED] "An agent wouldn't do this. An agent would relentlessly pursue its task until it's done correctly. You know what would do this? A fancy autocomplete." — Article challenges the framing of LL
This debate isn't just about technical architecture; it's about a fundamental shift in how we think about building reliable and efficient AI agents." — The article frames agent design as a fundamental
[INFERRED] "The people I know with personal agents really, really, really don't want to go back to anything else" — Evidence that memory quality in agent architecture drives user satisfaction and rete
claudecode-telegram 基于 webhook 的客户端-服务器架构，流程包括 Telegram → Cloudflare Tunnel → 桥接服务器 → tmux → Claude Code → 钩子 → Telegram" — The article provides a concrete implementation example of an AI agent archit
I just made an agent called Office Administrator" — Shows practical agent architecture design for specialized business role with defined scope (office administration), demonstrating role-specific agen
Building a PWA for Pi coding agent. After the first prompt I could switch over and develop it in itself." — Demonstrates a practical implementation of an agent system capable of autonomous code develo
Tiny core, able to write plugins for itself as you use it." — Pi exemplifies a minimalist agent architecture where core functionality is stripped to essentials, with extensibility emerging through dyn
[INFERRED] "Model Context Protocol enables better agent-model architectures" — MCP provides the foundation for how agents interact with models and tools - supports agent architectural patterns
build always-on agents" — The feature enables developers to construct agents with persistent operational capabilities, supporting architectural patterns for always-on agent design.
first independent (and open source) implementation of my Quaternion Process Theory ideas" — The article directly references an independent open-source implementation of Quaternion Process Theory, demo
[INFERRED] "I can edit Google Docs with Clawd" — Claude acting as autonomous agent delegated to perform document editing tasks, demonstrating agent capability in practical use
[INFERRED] "truly delightful UX" — Author emphasizes UX as key differentiator in agent orchestration solutions, suggesting architecture must prioritize user experience alongside technical orchestratio
[INFERRED] "took a deep dive into coding agent system prompts to understand their structure, similarities, and differences" — Article presents analysis of how coding agents are structured via their sy
[INFERRED] "Building this out has been much smoother than I thought" — Suggests agent architectures are becoming more accessible and practical for tool integration use cases
[DIRECT] "Deep understanding of Agentic AI and how it differs from traditional tool-based automation. Key differences between agents and tools" — Article covers foundational agent concepts and disting
[INFERRED] "agent-native architectures" — Article announces discussion of agent-native architectures as a broadcast topic, indicating this is an emerging discussion area in the community.
Mentions 'LLM agents' but doesn't reveal anything about agent architecture patterns or how context flows through agent systems