Building effective AI agents
Anthropic
Build the smallest system that works: Anthropic argues that added agent complexity is justified only when evaluation shows it improves outcomes.
Open library
The library
The essential reading of generative AI and agents, open in a clean reading view, with Meow and Standard summaries when available. No account needed to read.
560 papers · 558 with a Meow summary · 333 posts
The selected papers and blogs: the essential papers of generative AI and agents, and the key lab publications and essays, each with Meow and Standard summaries.
Effective context engineering for AI agents
Anthropic
Anthropic’s central shift: agent reliability depends on curating the full, changing context at each turn—not just polishing a prompt—because longer contexts can reduce recall and focus.
Writing effective tools for AI agents
Anthropic
Tools for agents should be designed around variable agent behavior and limited context, then improved through realistic evaluation—not treated as ordinary API wrappers.
How we built our multi-agent research system
Anthropic
Open-ended research is path-dependent: what you search next often depends on what you just found.
Harness design for long-running apps
Anthropic
A multi-agent harness made long app-building runs more capable, but its lasting lesson is selective: keep only the planning, decomposition, and evaluation the current model still needs.
Equipping agents with Agent Skills
Anthropic
Agent Skills package instructions, scripts, and resources for agents to load as needed. An update says they have been published as an open standard for cross-platform portability.
Making Claude Code safer with sandboxing
Anthropic
Claude Code’s sandbox gives its coding agent room to run commands inside set file and network boundaries, aiming to reduce permission prompts while limiting harm from prompt injection.
A postmortem of three recent issues
Anthropic
Anthropic says three infrastructure bugs—not demand, time of day, or server load—intermittently degraded Claude responses from August into early September.
Building a C compiler with parallel Claudes
Anthropic
The kernel became parallelizable only after a GCC-based oracle helped agents isolate different failing files.
Measuring AI agent autonomy in practice
Anthropic
In practice, an agent’s autonomy is not just a model capability: the model, the user, and the product shape it together.
Trustworthy agents in practice
Anthropic
Anthropic’s main claim: agent safety depends on the whole system, not just the model. Autonomy brings useful capabilities but also more room for mistaken actions and prompt-injection attacks.
Mitigating prompt injections in browser use
Anthropic
Claude Opus 4.5 and improved safeguards sharply reduced browser prompt-injection success, but the reported gains do not solve the problem.
Scaling Managed Agents with stable interfaces
Anthropic
Managed Agents separates the agent’s session, harness, and execution tools behind interfaces intended to stay stable as models and implementations change.
How we built Claude Code auto mode
Anthropic
Claude Code’s auto mode aims to reduce approval fatigue without switching permissions off: it uses classifiers to allow higher autonomy while retaining safeguards.
Infrastructure noise in agentic coding evals
Anthropic
Agentic coding benchmark scores can shift with the machine budget, not just model capability, so a small leaderboard lead may measure infrastructure as much as coding.
Introducing the Model Context Protocol
Anthropic
MCP replaces separate AI-to-data connectors with an open protocol that gives developers one shared way to connect tools and data.
Developing a computer use model
Anthropic
In its Aug. 28, 2025 update, Anthropic said Claude 3.5 Sonnet could operate ordinary software through screen, cursor, and keyboard actions—moving AI beyond custom tools, though the capability was still in public beta.
Gemini CLI: your open-source AI agent
Taylor Mullen · Google
Gemini CLI brings Gemini into developers’ terminals, with high—but explicitly capped—free-preview usage for individuals.
Announcing the Agent2Agent Protocol (A2A)- Google Developers Blog
Rao Surapaneni · Google
A2A aims to make agents from different vendors and frameworks collaborate across enterprise systems through a shared protocol.
Qwen3-Coder: Agentic Coding in the World
Qwen Team · Qwen
Qwen3-Coder combines a long-context coding model with training that uses execution checks and multi-step interaction—the capabilities the announcement emphasizes for agentic work.
QwenLM/Qwen-Agent: Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.
Qwen
Qwen-Agent turns Qwen’s instruction following and tool use into a framework for building LLM applications, not just ready-made chat demos.
GPT-4.1 Prompting Guide
OpenAI
GPT-4.1 follows prompts more literally than its predecessors, so developers should make desired behavior explicit and expect to migrate prompts that relied on implied intent.
What is the Model Context Protocol (MCP)?
Anthropic
MCP is an open standard for connecting AI applications to external systems.
Harness Engineering for Self-Improvement
Lilian Weng
The article’s key shift is from improving an agent’s answer to improving the machinery that repeatedly produces and checks answers.
Prompt Engineering
Lilian Weng
Prompt engineering steers a language model without changing its weights, but it is empirical: a prompt that works for one model may fail on another.
Designing agentic loops
Simon Willison
The key design problem is choosing goals, tools, and boundaries that let an agent iterate toward a solution without making its freedom the biggest risk.
Vibe engineering
Simon Willison
Willison proposes “vibe engineering” for AI-assisted production work where engineers remain accountable—not for accepting code just because it appears to work.
Here’s how I use LLMs to help me write code
Simon Willison
Treat an LLM as a fast, fallible coding partner: the developer must supply direction, verify behavior, and take over when judgment matters.
Building A Generative AI Platform
Chip Huyen
A generative AI platform is more than a model call: it is a set of optional components that can be added as an application’s needs grow.
Building LLM applications for production
Chip Huyen
A working LLM demo is not proof of a reliable production system: prompts can fail silently, and multi-step workflows add more ways to go wrong.
Common pitfalls when building generative AI applications
Chip Huyen
A working GenAI demo proves possibility—not that a model is the simplest tool, that users want it, or that it is ready for production.
Patterns for Building LLM-based Systems & Products
Eugene Yan
An LLM demo is not a dependable product: the guide argues for measuring the real task and combining retrieval, specialization, safeguards, and user-centered design.
What We’ve Learned From A Year of Building with LLMs
Eugene Yan · Applied LLMs
The authors’ year of building yields a practical thesis: getting an LLM demo to work is easier than making a dependable product; reliability comes from engineering the system around the model.
Introducing smolagents: simple agents that write actions in code.
Aymeric Roucher · Hugging Face
smolagents launches a compact library for language models to control tool-using workflows, with agents that write actions in code at its center.
Context Engineering
LangChain Accounts · LangChain
An agent’s context window is working memory, not a transcript that should grow without limit.
Deep Agents
LangChain Accounts · LangChain
A plain LLM tool loop can be “shallow”—unable to plan across longer tasks. The proposed fix keeps the loop and adds scaffolding for planning, subagents, and context.
How to think about agent frameworks
LangChain Accounts · LangChain
Agent frameworks are easy to compare badly: this article separates workflow-versus-agent behavior from orchestration style and high-level abstractions, with context control as the reliability test.
Introducing ambient agents
LangChain Accounts · LangChain
Ambient agents move beyond one-at-a-time, user-started chat: they listen for events, can act across several at once, and still bring people in when needed.
Don’t Build Multi-Agents
Walden Yan · Cognition
The failure isn’t simply parallelism: it’s agents making decisions without seeing the context and choices that shaped the rest of the work.
Context Engineering for AI Agents: Lessons from Building Manus
Manus
Manus chose context engineering as a way to improve its agent in hours instead of weeks, without tying the product to the underlying models.