Agents in Macro sit on top of unified memory, so one prompt can reach your email, messages, tasks, docs, calls, and connectors. These recipes are starting points: paste them into an agent chat (c + a), an automation, a channel via @Macro, or any MCP client connected to your workspace.
Daily inbox brief
Create an automation (agents module → Automations tab → Create an automation) scheduled for each morning:
The result lands in your inbox at the scheduled time, every day.
Project status on demand
Ask in any agent chat. You don’t need to point it at anything, since memory is refreshed from your team’s activity:
You can @mention a specific channel, doc, or task in the prompt to skip the search and guarantee the right context. See Mentions in Agents.
Weekly team recap
An automation for Friday afternoons:
Turn a call into tasks
After a call ends, its transcript is already in your workspace. From the call (or any chat):
Draft replies in your voice
Review the draft, then tell the agent to send it, or send it yourself from the composer.
Answer questions inside a channel
Mention @Macro in any channel to pull an agent into the conversation, with the channel as context:
Connect the tool under Settings → Connectors, then:
See Switch to Macro for tool-by-tool migration guides.
Use your workspace from Claude Code
Once you’ve connected the Macro MCP server, your coding agent has the same reach. It can search your workspace, read threads and docs, create documents, send email, and update task properties with the MCP tools:
Agents inherit your permissions: they can only see what you can see, and anything they create or send is attributed to you. Start prompts with “draft, don’t send” until you’re comfortable letting them send.
