Harness design for long-running application development
… Why naive implementations fall short We've previously shown that harness design has a substantial impact on the effectiveness of long running agentic coding. …
Search
Trustworthy agents in practice
… We go into greater technical detail on this topic in our submission to NIST's Center for AI Standards and Innovation CAISI on agentic security. …
Long-running Claude for scientific computing
Science Long-running Claude for scientific computing Mar 23, 2026 In this post, Siddharth Mishra-Sharma , a researcher on the Discovery team, explains how to apply multi-day agentic coding workflows—test oracles, persistent memory, and orchestration patterns—to scientific computing tasks even outsi… …
Agents for financial services
… Carlyle has adopted Claude as a key part of our AI technology stack because of its strong coding capabilities, agentic reasoning, and continual advances in both the underlying models and key features. …
Claude Code auto mode: a safer way to skip permissions
… Bypassing permissions is zero-maintenance but offers no protection. Manual prompts sit in the middle, and in practice users accept 93% of them anyway. We keep an internal incident log focused on agentic misbehaviors. …
Introducing Claude Opus 4.5
… A common benchmark for agentic capabilities is τ2-bench , which measures the performance of agents in real-world, multi-turn tasks. In one scenario, models have to act as an airline service agent helping a distressed customer. …
Focus areas for The Anthropic Institute
… Sharing the gains: What pre- or re-distributive mechanisms could effectively spread the gains from AI development and deployment more broadly? Transaction costs in markets: How does AI affect systems of exchange and transaction costs in marketplaces? …
Building Effective AI Agents
… Agents , on the other hand, are systems where LLMs dynamically direct their own processes and tool usage, maintaining control over how they accomplish tasks. Below, we will explore both types of agentic systems in detail. …
Demystifying evals for AI agents
… Single-turn evaluations are straightforward: a prompt, a response, and grading logic. For earlier LLMs, single-turn, non-agentic evals were the main evaluation method. As AI capabilities have advanced, multi-turn evaluations have become increasingly common. …
Claude Opus 4.6
… It plans more carefully, sustains agentic tasks for longer, can operate more reliably in larger codebases, and has better code review and debugging skills to catch its own mistakes. …