LangChain: 'How to Give Your Agent Memory' — LangSmith Context Hub Pattern
LangChain's Jake Broekhuizen shows how to close the agent memory loop using LangSmith traces, Engine analysis, and Context Hub storage.
LangChain's Jake Broekhuizen shows how to close the agent memory loop using LangSmith traces, Engine analysis, and Context Hub storage.
OpenClaw 2026.6.11 merges 305 PRs with Slack relay mode, RAFT CLI wake bridge, Mattermost support, and Android improvements.
Pydantic AI v2.0 stable is out with a capability primitive that unifies tools, hooks, and instructions into one composable agent-building block.
Z.ai's GLM-5.2 ships as a 744B open-weights MIT model with 1M context for coding agents — deployable with OpenClaw, nearly matching Claude Opus 4.8 performance.
Claude Code v2.1.185 doubles stream-stall timeout to 20s, reducing false retries in long-running and multi-agent sessions.
FINOS releases the AIGF MCP Server, embedding AI governance and regulatory compliance (EU AI Act, DORA) directly into agentic MCP workflows.
Headroom, a Netflix engineer's open-source tool, compresses agent context locally by up to 95% — now trending with a dedicated OpenClaw plugin.
OpenClaw 2026.6.10-beta.1 ships reliable agent state, Zalo channel support, CLI dry-run previews, and richer Discord/Telegram delivery.
Step-by-step operator checklist for upgrading OpenClaw from 2026.6.8 to 2026.6.9: Telegram HTML delivery, Codex auto-approvals, and agent recovery settings.
Ponytail forces AI coding agents to ask 'does this need to exist?' before writing code — cutting generated output 80-94% via a six-rung laziness ladder.
Google, Microsoft, Cisco, and 10+ industry partners launch ARD — a federated open protocol letting AI agents discover tools like a search engine.
Microsoft Build 2026: Azure Functions now lets you define AI agents in .agent.md markdown files — serverless, event-driven, MCP-ready.
Jensen Huang and Boris Cherny declare loops — not prompts — the unit of agentic AI value. Loop engineering is officially mainstream in 2026.
Adaline Labs' guide to agentic harness engineering — the five layers every production self-improving AI agent needs to avoid quality decay.
MCP Enterprise-Managed Authorization hits stable: Okta SSO now provisions AI tool access org-wide, eliminating per-user OAuth friction for Claude and VS Code.
Learn to build production-grade AI agents using Vercel's Agent Stack: AI SDK, AI Gateway, eve framework, and Vercel Connect — announced at Vercel Ship 2026.
Learn how to harden your local MCP server against AutoJack-style attacks where malicious webpages hijack AI browsing agents to execute arbitrary host processes.
Step-by-step guide to enabling Claude Code Artifacts for your Team or Enterprise plan and creating live, shareable dashboards from your AI coding sessions.
Vercel's 'eve' is a Next.js-style open-source agent framework treating each agent as a filesystem directory with built-in durable execution.
ctx goes GPL-3.0 open source — a hackable desktop workbench for coding agents with worktrees, merge queues, and transcript review built in Rust and Tauri.
GitHub's team slashed agent token costs up to 62% by pruning MCP tools and replacing LLM calls with direct CLI. A replicable engineering playbook for any agentic stack.
Databricks turns Agent Bricks into a full production platform at DAIS 2026 — covering model choice, data context, and governance for 100k+ enterprise agents.
OpenClaw 2026.6.8 arrives with richer Telegram/WhatsApp delivery, GLM-5.2 and Claude Haiku 4.5 support, stabler agent runs, and smarter search defaults.
Anthropic requires government ID and selfie for Claude Free/Pro/Max users from July 8. Enterprise and API accounts exempt. Processed by Persona Identities.
AWS Summit New York 2026 delivers Kiro spec-driven IDE, Bedrock AgentCore enterprise platform, and Amazon Quick. Here's what builders need to know.