Managed DeepAgents enters public beta with evals, memory, OAuth and Slack/GitHub integrations
hwchase17 · x · 2026-08-04
Move the summary for managed deepagents to public beta this week.
The focus is on the “boring” infrastructure around agents so users can spend time on agent logic instead. The listed pieces so far are:
- opinionated evals setup via Harbor
- agent- and user-level memory
- proper OAuth for tool access
- easy integrations with Slack and GitHub
- seamless sandbox integration
The post also asks what else would be useful in this stack.
Related event: LangChain's Managed Deep Agents enters public beta(2 posts)→
More from coding & agent
- Codex uses Blender and Unity to build an otter-mech game with full assets — emollick · 2026-08-04
- GitHub repo curates 100+ open-source AI agents, skills, and RAG apps — Saboo_Shubham_ · 2026-08-04
- Swarms Cloud adds saved workflows, session persistence, and 1,500+ models — KyeGomezB · 2026-08-04
- AI Coding Sparks a New 'Middle-Class' Software Consulting Market — manosaie · 2026-08-04
- OpenAI will retire GPT-5.4 from Codex on August 31 — ivan_bezdomny · 2026-08-04
- A verification skill every agent needs: computer and browser use — vikvang1 · 2026-08-04