Work
Work is Coworker's default AI chat mode for complex, multi-step research and analysis. Unlike a simple Q&A, Deep Work plans and executes tasks that require pulling from many data sources at once — meetings, documents, Slack, email, Jira, GitHub, CRM, and more.
Best for:
Generating comprehensive reports and strategic analyses
Synthesizing information across multiple tools and time periods
Taking action — creating Jira tickets, updating Notion pages, drafting Google Docs
Any task that requires reasoning across multiple data sources before producing an answer
Agents
Agents are autonomous AI workers you configure once and run automatically. Using Coworker's no-code Agent Builder, you write plain-language instructions, connect the agent to the tools and data sources it needs, and set a schedule or trigger. The agent then runs on its own and delivers results to you via Slack, email, or the Coworker interface — no manual effort required.
Best for:
Recurring tasks you do manually more than once a week (digests, status reports, pipeline scans)
Monitoring workflows — surface at-risk deals, flagging open tickets, tracking project health
Automating document or inbox triage
Any task triggered by an external event via API or Zapier/Make
Agents can be triggered manually, on a schedule (e.g., every Monday at 9am), or via the Coworker API.
Notetaker & Recorder
Coworker Notetaker runs on the background using the Coworker desktop app. It generates the transcript and produces a summary report. No bots needed.
Coworker Recorder is a bot that joins your Google Meet or Zoom meetings to capture audio, generate a transcript, and produce a summary report. It appears as an attendee named Coworker Recorder and can join automatically or on demand.
What you get after every meeting:
Transcript — available approximately 30 seconds after the call ends
Summary report — including discussion topics and action items, available approximately 1–2 minutes after the call ends
Transcripts and summaries are searchable within Coworker and used to power Work and Customer Intelligence queries about past meetings.
Customer Intelligence
Customer Intelligence (CI) is a dedicated add-on module that gives customer-facing teams a single, unified view of every account. It combines your CRM data with AI-generated insights drawn from all of Coworker's connected data sources — meetings, Slack, email, documents, and more.
The core of CI is the Magic Table: an AI-enriched data table with three types of columns:
CRM columns — Synced automatically from HubSpot or Salesforce (e.g., ARR, renewal date, deal stage)
AI Insight columns — AI-generated answers to questions you define about each account, pulling context from all connected data sources
Text columns — Custom fields for manual notes and additional context
Best for: Pre-call prep, renewal management, QBR preparation, and team handoffs.
Note: Customer Intelligence is a separate add-on. If you don't see it in your left-hand navigation, contact your Coworker admin or account team to request access.
CoDrive
CoDrive is your team's shared workspace inside Coworker for storing, organizing, and accessing the work you create — artifacts, agents, skills, and meetings — all in one place. No more hunting across individual drives or losing track of shared work.
What's in CoDrive
Artifacts (documents, decks, canvases, and other generated content)
Agents you've built or have access to
Skills shared with you or your organization
Meetings, automatically organized into folders by day
Coworker MCP
Coworker MCP server supercharges your third-party AI apps with company context via Organizational Memory.
MCP stands for Model Context Protocol — an open standard that allows external AI agents and tools to connect to Coworker using a standardized interface. Rather than requiring custom integrations for each AI client or workflow, Coworker's MCP server exposes Coworker's capabilities natively, so any MCP-compatible agent can discover and use them automatically.