Anthropic and OpenAI release AI agent development guide
The 115-article collection covers the full lifecycle of agent systems, from design to production. The guide systematizes approaches to building agents, focusing on the Model Context Protocol (MCP) standard for integrating external data and tools.
The guide details architectural patterns for workflow orchestration, task handover, security, and autonomy control. It also covers tools like Claude Code and OpenAI Codex, including testing with Evals, context management, sandboxing for secure code execution, and creating agent self-improvement loops.
Up next
Anthropic to temporarily limit access to Claude Fable 5 for subscribers
Anthropic is pulling Claude Fable 5 from standard subscriptions after June 23 due to capacity constraints. The model will still be available via API at $10/m input and $50/m output. For now, subscribers will need to buy separate credits to access Fable 5.
Related stories
- Claude memory: say what it’s for, let dreaming fix notes
- Claude Fable 5 tripled Opus 4.6’s EVE Online trading gains
- Cat Wu, Claude Code’s product lead, joins OpenAI
- Irregular CEO blames human error for AI agent breaches
- Anthropic joins a shared robots.txt opt-out for AI training
- Intelligence Index v5: Opus 5.5 at 58, GPT-6 Astra at 57
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.