AI News Summary - October 27, 2026

Claude Code Introduces Multi-Repository Context Awareness

Anthropic's Claude Code now supports simultaneous context from up to 10 linked repositories, enabling developers to work across microservices architectures more efficiently. The update includes automatic dependency mapping and cross-repo refactoring suggestions. Early adopters report 40% faster development times for complex distributed systems.

MCP 2.0 Specification Released with Bidirectional Streaming

The Model Context Protocol consortium announced MCP 2.0, featuring bidirectional streaming capabilities that allow real-time context updates between AI models and external data sources. The update includes improved authentication mechanisms and native support for vector databases. Over 50 major AI tools have already committed to implementing the new specification.

Open-Source Agent Framework "Cascade" Reaches 1.0

The Cascade framework for building autonomous AI agents hit version 1.0, offering built-in memory persistence, tool composition, and multi-agent coordination. The project includes pre-built connectors for popular APIs and databases, plus a visual workflow editor for non-programmers. The framework has been downloaded over 100,000 times since its beta launch in August.

GitHub Copilot Workspace Adds Team Knowledge Graphs

GitHub unveiled team-level knowledge graphs in Copilot Workspace that automatically map institutional knowledge from past pull requests, documentation, and code reviews. The feature helps new team members onboard faster and suggests context-aware solutions based on team-specific patterns. Enterprise customers will gain access in a phased rollout starting next week.