CodeAct

Microsoft shipped the Agent Framework Harness on July 22, a production-ready execution layer that turns an LLM into an autonomous agent with automatic tool-calling, planning, persistent memory, contex…

Model routing in agentic systems is far more complex than a simple classification problem, according to an enterprise article. Across 417 tasks on the AppWorld Test Challenge, Claude Sonnet 4.6 cost $…

Researchers introduced SPEAR (Sandboxed Prompt Engineer with Active Roll-back), a code-augmented agentic optimizer that autonomously rewrites prompts using a Python sandbox for structural error analys…