Top Story
[Release] anthropics/claude-code v2.1.118: v2.1.118
Claude Code v2.1.118 introduces vim visual mode selection, consolidates `/cost` and `/stats` into `/usage`, enables custom themes with JSON editing, and allows hooks to invoke MCP tools directly. The release includes significant improvements to MCP OAuth authentication, auto mode configuration with defaults merging, and numerous bug fixes across WSL integration, plugin management, credential handling, and remote sessions.
- Added vim visual mode (`v`) and visual-line mode (`V`) with selection operators and visual feedback for enhanced text editing
- Merged `/cost` and `/stats` commands into unified `/usage` command while maintaining backward compatibility as typing shortcuts
- Implemented custom theme creation and switching via `/theme` command with JSON file editing in `~/.claude/themes/` and plugin theme distribution support
- +7 more key points...
By Topic
Agent Teams
See all 7 →[Release] google/adk-python v2.0.0b1: v2.0.0b1
Google ADK Python v2.0.0-beta.1 marks the transition to Beta phase with significant architectural improvements. Key additions include a full Workflow graph orchestration system with NodeRunner for execution isolation, explicit ReAct loop nodes for agent execution, and Human-in-the-loop resumption capabilities. The release also addresses a critical RCE vulnerability in YAML configuration handling and optimizes performance for single-turn LLM agents.
sasha-gitg
Build AI Agents That Use Tools 🤖 | AutoGen register_function Explained
This tutorial explains how to build AI agents using AutoGen's register_function feature, enabling agents to use external tools and functions. The course covers practical implementation of tool-calling capabilities, allowing agents to extend their functionality beyond language model limitations. The GitHub repository provides complete code examples for building production-ready AI agents with integrated tool support.
Nidhi Chouhan
Gemini Enterprise Agent Platform, Google's New Agent Stack Explained
Google has launched the Gemini Enterprise Agent Platform, a comprehensive stack for building and deploying enterprise AI agents. The platform integrates Gemini models with enterprise tools, APIs, and data sources to enable organizations to create autonomous agents that can perform complex business tasks. This represents Google's strategic push into the enterprise AI agent market with production-ready infrastructure.
TechWealth Hub
Skills & Tools
[Release] openai/openai-agents-python v0.14.5: v0.14.5
OpenAI Agents Python v0.14.5 release includes three key fixes and features: Modal sandbox idle timeout configuration, HITL resume tool output serving, and streamed terminal output backfilling. Documentation improvements clarify lazy skill source host paths. This patch release addresses sandbox management, human-in-the-loop workflows, and output streaming reliability.
seratch
OpenCLAW vs GitHub Copilot: Which AI Coder Reigns Supreme?
This video compares OpenCLAW and GitHub Copilot, two AI coding assistants, challenging common assumptions about their differences. The analysis examines their core capabilities, use cases, and performance characteristics to determine which tool better serves different developer needs. The comparison reveals that the 'right' choice depends on specific workflow requirements rather than one tool being universally superior.
Tech Trouble Shooters
How to Run Your First Task in Manus AI | Beginner Step-by-Step Tutorial
This beginner tutorial guides users through running their first task in Manus AI, a platform for AI agent development. The video provides step-by-step instructions to set up and execute an initial task, helping new users understand the basic workflow and interface. It includes practical demonstrations and tips for getting started with the platform's core features.
EProducts Empire
Security
[Release] crewaiInc/crewAI 1.14.3a3: 1.14.3a3
crewAI 1.14.3a3 introduces e2b support and Azure credential fallback functionality while addressing a critical security vulnerability in lxml. The release achieves a ~29% improvement in cold start time through lazy-loading optimizations of MCP SDK and event types. Documentation updates include removal of pricing FAQs across all locales.
greysonlalonde
OpenClaw Trojan AI agents give attackers remote control over 28,000 systems
SecurityScorecard discovered over 28,000 OpenClaw AI agent systems exposed online with inadequate security controls, affecting approximately 63% of identified instances. These systems lack proper safeguards, creating significant vulnerability to remote exploitation and unauthorized access. The exposure represents a critical security risk in AI agent deployments, highlighting the urgent need for stronger authentication, access controls, and monitoring mechanisms in production AI systems.
Saeb Masarwa Eng
Integrations
Today's Breakdown
Sources
Content Types
Difficulty
Want to save articles & build playbooks?
Create a free account to save articles to your library, build AI-generated implementation playbooks, and get content matched to your stack.
Or just get it by email
No account needed. Unsubscribe anytime.