Keep chatting with your main agent while a long research or browser job runs in the background: mark that subagent async so it reports back when finished instead of freezing the conversation: AI agent prompt
LangChain's Oct 29, 2026 Managed Deep Agents v0.10 post introduced asynchronous subagents: add async: True on a subagent definition so the orchestrator starts that helper on its own thread and keeps talking with you while the work runs. The agent can check, update or cancel the background task; when the subagent finishes, results arrive mid-run if the orchestrator is busy, or wake an idle agent with a new run, without interrupting a pending human approval. The same release adds browser use with browser_handoff so the agent can work on sites with no API and hand you the window at a login wall so passwords never enter model messages. Product lead Sydney Runkle compared the pattern to a chief of staff who delegates without disappearing. Useful for Slack-based agents that must answer follow-ups while research or multi-step browser work continues.
The prompt
Design an orchestrator agent for [my product / Slack bot] that stays responsive while long jobs run. Jobs that should be async: [research, multi-page browser task, report build]. Rules: never block the main chat on those jobs; never put passwords in model context—hand the browser to me at login; escalate if a background job fails twice. 1) List which subagents are sync vs async and why. 2) Draft the subagent definitions (name, description, system prompt, async flag) in plain language I can paste into Managed Deep Agents or recreate in my stack. 3) Write what the main agent says when it starts, checks on, and receives a finished background job. 4) Give a 5-case test plan (user asks a follow-up mid-research; login wall; cancel; approval pending; double failure). Do not deploy anything in this chat.
Fill in the blanks, then copy
This prompt has 2 [bracket]s to fill in. Type your details below and the prompt updates as you go. Your answers stay in this browser and are reused on other prompts with the same blanks. Nothing is sent to us.
Example values are fictional, just to show how a filled-in prompt reads. Replace them with your real details.How to use it
- Copy the prompt above.
- Replace every [bracket] with your own details (use the fill-in-the-blanks box above, which remembers your answers for the next prompt).
- Paste it into LangChain Managed Deep Agents, ChatGPT, Claude and review the answer before you use it.
Works with: LangChain Managed Deep Agents, ChatGPT, Claude, Slack
Real-world example: langchain.com/blog/managed-deep-agents-browser-use-and-asynchronous-su
Featured in: ChatGPT prompts for business email, replies and follow-ups · AI prompts for writing SOPs, reports, policies and checklists
Before you use it: what to check
Already built into this prompt
- Rules: never block the main chat on those jobs; never put passwords in model context—hand the browser to me at login; escalate if a background job fails twice.
- Do not deploy anything in this chat.
Check yourself
- Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings.
- Read and edit the draft before it reaches a customer or goes public. If you later automate this prompt, start with "draft and notify me", not "send automatically".
- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
Using an AI agent? Download this prompt as a SKILL.md agent skill with these checks built in, or get the .zip for Claude.ai. How to install skills.
Want to make this prompt even better? Grade it with the free prompt grader or build your own with the AI prompt generator.
32 copy-paste agent prompts (PDF) plus 5 new prompts every week.
Get it free