LangChain launches Managed Deep Agents: sandbox proxies make production agents secure by default
LangChain · x · 2026-10-17
LangChain released Managed Deep Agents (MDA), pitched as the simplest way to build and deploy production agents: you define the agent's intelligence—instructions, skills, tools, and model—while MDA runs everything underneath, with no servers or infrastructure to manage.
Key pieces:
- The Deep Agents harness: an agent loop that plans, calls tools, manages a filesystem, and delegates to subagents
- A managed runtime on LangSmith Agent Server providing API, threads, runs, streaming, and an MCP endpoint
Agents are organized as a project folder: agent.py for model/tools/middleware, instructions.md for behavior, skills/ for reusable skills.
On security, MDA addresses the difficulty of securing production agents by using sandbox proxies to keep credentials safe, making shipped agents secure by default.
Related event: LangChain Launches Managed Deep Agents for Production-Grade AI Agents(2 posts)→
More from coding & agent
- "AI hasn't solved coding": easy tasks hide the bumps, warns dev in Anthropic debate — gerardsans · 2026-10-17
- Interactive 3D knee biomechanics sandbox built with Gemini 4 Argon and three.js — creatoroff · 2026-10-17
- Context Inertia: when LLM agents latch onto a wrong hypothesis and won't let go — rseroter · 2026-10-17
- iOS Devs: Install These Agent Skills to Make Your App Designs Stand Out — stefanjblos · 2026-10-17
- Devin Can Now Build and Test iPhone Duo Apps in Cloud VMs with Video Verification — DevinAI · 2026-10-17
- Reddit: getting agents to reliably update CRM records after meetings — Tough_Adeptness_6160 · 2026-10-17