AI News

The latest from the AI and MCP ecosystem, curated daily.

Anthropic EngineeringClear

Claude Code auto mode: a safer way to skip permissions

Anthropic EngineeringMar 25, 2026

Claude Code auto mode: a safer way to skip permissions

Anthropic introduced Auto mode for Claude Code, which lets it execute commands and edit files autonomously without prompting for permission at each step — but with sandboxing guardrails that make it safe to do so. The mode intelligently decides when to proceed vs. pause based on risk level, reducing friction for long-running coding tasks. A meaningful step toward truly autonomous software agents.

claude-codenew-featureagentsautonomous

Harness design for long-running application development

Anthropic EngineeringMar 24, 2026

Harness design for long-running application development

A deep technical post on how to architect agent harnesses that stay reliable across long development sessions — covering state management, context window hygiene, error recovery, and how to structure tool calls so agents don't drift. Anthropic shares patterns from building Claude Code and their internal agent infrastructure. Essential for teams building production-grade agentic systems.

agentsagent-harnessdeveloper-toolsclaude-code

Quantifying infrastructure noise in agentic coding evals

Anthropic EngineeringMar 11, 2026

Quantifying infrastructure noise in agentic coding evals

Anthropic researchers found that infrastructure configuration — things like CPU throttling, network latency, and VM scheduling — can swing agentic coding benchmark scores by several percentage points, sometimes more than the actual gap between top models on leaderboards. This means many published eval comparisons are noisier than they appear, and teams running their own benchmarks need to control for environment carefully. Important reading for anyone doing serious agent evaluation work.

agentsevalsbenchmarksresearch

Eval awareness in Claude Opus 4.6's BrowseComp performance

Anthropic EngineeringMar 6, 2026

Eval awareness in Claude Opus 4.6's BrowseComp performance

Anthropic investigated whether Claude Opus 4.6 behaves differently when it detects it's being evaluated, specifically on the BrowseComp web-browsing benchmark. The findings have implications for how much trust to place in benchmark scores vs. real-world performance. Relevant for anyone relying on published evals to make model selection decisions.

researchevalsclaudebenchmarks

Building a C compiler with a team of parallel Claudes

Anthropic EngineeringFeb 5, 2026

Building a C compiler with a team of parallel Claudes

Anthropic engineers built a working C compiler by running multiple Claude instances in parallel, each responsible for different compiler stages, with a coordinator orchestrating the workflow. The experiment demonstrates real multi-agent software engineering at a non-trivial scale and surfaces practical lessons about task decomposition, inter-agent communication, and error recovery. A compelling proof-of-concept for parallel agent coding systems.

agentsclaude-codemulti-agentcoding

Designing AI-resistant technical evaluations

Anthropic EngineeringJan 21, 2026

Designing AI-resistant technical evaluations

Anthropic shares what they learned from three iterations of a performance engineering take-home assessment that Claude kept solving correctly — forcing them to redesign it to remain useful for hiring. The post is a practical guide to crafting technical evals that test genuine human reasoning rather than tasks AI can trivially handle. Useful for any team still using take-home coding challenges.

researchevalsai-codinghiring

Demystifying evals for AI agents

Anthropic EngineeringJan 9, 2026

Demystifying evals for AI agents

A practical breakdown of how to design, run, and interpret evaluations specifically for agentic systems — covering trajectory evals, outcome evals, LLM-as-judge approaches, and the unique challenges agents pose vs. static model benchmarks. Anthropic draws on their experience evaluating Claude Code and research agents. One of the most useful practical guides on agent evals published to date.

agentsevalsresearchdeveloper-tools

Effective harnesses for long-running agents

Anthropic EngineeringNov 26, 2025

Effective harnesses for long-running agents

Anthropic outlines architectural patterns for building agent harnesses that remain reliable over extended runs — covering context management, checkpointing, graceful error recovery, and how to structure prompts so agents stay on track through dozens of tool calls. Drawn from experience building Claude Code and internal research agents. Pairs well with their later harness design post.

agentsagent-harnessdeveloper-toolspatterns

Introducing advanced tool use on the Claude Developer Platform

Anthropic EngineeringNov 24, 2025

Introducing advanced tool use on the Claude Developer Platform

Anthropic launched advanced tool use capabilities on the Claude API — enabling agents to dynamically discover, learn, and execute tools at runtime rather than having a fixed toolset hardcoded at startup. This allows more flexible agent architectures where Claude can adapt its tool usage based on task context. A significant API capability upgrade for developers building production agents.

claudeapinew-featureagents

Code execution with MCP: Building more efficient agents

Anthropic EngineeringNov 4, 2025

Code execution with MCP: Building more efficient agents

Anthropic demonstrates how combining MCP with code execution lets agents handle far more tools while consuming dramatically fewer tokens — up to 98.7% reduction in context overhead in their tests. Instead of passing tool results back through the context window, agents write code to chain tool calls and filter outputs before they reach the model. A major efficiency win for complex multi-tool agent pipelines.

mcpagentscode-executionefficiency

Beyond permission prompts: making Claude Code more secure and autonomous

Anthropic EngineeringOct 20, 2025

Beyond permission prompts: making Claude Code more secure and autonomous

Claude Code gained filesystem and network sandboxing, letting it run autonomously with fewer permission interruptions while remaining safe. The sandboxing isolates risky operations so Claude can proceed confidently without asking the user about every file write or network call. A key step in making agentic coding tools both safer and less annoying to use.

claude-codenew-featuresecurityautonomous

Effective context engineering for AI agents

Anthropic EngineeringSep 29, 2025

Effective context engineering for AI agents

Anthropic shares practical techniques for managing what goes into an agent's context window — what to include, what to summarise, what to drop, and how to structure information so agents reason more reliably. Context engineering is increasingly recognised as a core skill for agent developers, and this is one of the clearest practical guides on the topic.

agentscontext-engineeringdeveloper-toolspatterns

A postmortem of three recent issues

Anthropic EngineeringSep 17, 2025

A postmortem of three recent issues

Anthropic published a transparent technical postmortem on three bugs that intermittently degraded Claude's responses — explaining what happened, why each took time to diagnose and fix, and what engineering changes resulted. Rare public transparency from an AI lab on production reliability issues. Useful reading for teams thinking about AI system reliability.

anthropicreliabilityengineeringpostmortem

Writing effective tools for agents — with agents

Anthropic EngineeringSep 11, 2025

Writing effective tools for agents — with agents

Anthropic used Claude agents to help design and iterate on tool definitions for other agents — effectively dog-fooding agent tooling. The post covers what makes a tool definition good (clear name, typed params, explicit error cases) and how using agents to write tools surfaces issues that humans miss. Meta but practical.

agentstoolsdeveloper-toolspatterns

Desktop Extensions: One-click MCP server installation for Claude Desktop

Anthropic EngineeringJun 26, 2025

Desktop Extensions: One-click MCP server installation for Claude Desktop

Anthropic launched Desktop Extensions — a one-click installation system for MCP servers directly in Claude Desktop, eliminating the need to manually edit JSON config files. Users can browse, install, and manage MCP servers from a UI without touching the terminal. A major UX improvement that significantly lowers the barrier to MCP adoption.

mcpclaudenew-featuredeveloper-tools

How we built our multi-agent research system

Anthropic EngineeringJun 13, 2025

How we built our multi-agent research system

Anthropic walks through the engineering decisions behind Claude's Research feature — a multi-agent system that parallelises web search, synthesises findings across agents, and produces long-form research reports. The post covers orchestration architecture, reliability challenges, and how they handled conflicting information between sub-agents. Valuable real-world reference for anyone building research or RAG-style multi-agent systems.

multi-agentagentsresearchclaude

Claude Code: Best practices for agentic coding

Anthropic EngineeringApr 18, 2025

Claude Code: Best practices for agentic coding

Anthropic's official best practices guide for Claude Code — covering how to structure tasks, write effective CLAUDE.md files, use tools efficiently, manage context, and get reliable results on complex coding tasks. The canonical reference for getting the most out of Claude Code in production. Should be required reading before any serious Claude Code deployment.

claude-codedeveloper-toolsbest-practicescoding

The "think" tool: Enabling Claude to stop and think in complex tool use situations

Anthropic EngineeringMar 20, 2025

The "think" tool: Enabling Claude to stop and think in complex tool use situations

Anthropic introduced a dedicated "think" tool that gives Claude a scratchpad to reason through complex multi-step tool use situations before acting. In tests, adding the think tool significantly improved accuracy on tasks requiring careful sequencing of tool calls. A simple but effective technique any agent developer can adopt immediately.

agentstoolsclaudereasoning

Raising the bar on SWE-bench Verified with Claude 3.5 Sonnet

Anthropic EngineeringJan 6, 2025

Raising the bar on SWE-bench Verified with Claude 3.5 Sonnet

Anthropic achieved then-state-of-the-art performance on SWE-bench Verified with Claude 3.5 Sonnet using only a bash tool and a text editor — no scaffolding tricks. The post details the technical approach and what the benchmark results imply about real-world software engineering capability. A landmark result that established Claude as a serious coding model.

claudebenchmarkscodingswe-bench

Building effective agents

Anthropic EngineeringDec 19, 2024

Building effective agents

Anthropic's foundational guide to building reliable AI agents — covering when to use agents vs. simpler pipelines, how to design tool interfaces, how to handle failures gracefully, and the core patterns (augmented LLM, routing, parallelisation, orchestrator-subagent) that underpin most production agent systems. Still one of the best starting points for anyone new to agent development.

agentsdeveloper-toolspatternsclaude