Meet the Agents: How Three AI Teammates Cover Your Entire Meeting Lifecycle
Three Agents, One Meeting Lifecycle
Most AI meeting tools give you a single assistant and call it done. We took a different approach. At botzone.ai, your team gets three specialised AI agents. Each one covers a different phase of the meeting lifecycle — from the discovery conversation to the daily standup. Together, they ensure nothing falls through the cracks.
Here is who they are, what they do, and why we built them this way.
Sage — Meeting Intelligence
Status: Live
Sage is the core of botzone.ai. It joins your Google Meet calls as a participant, announces itself, and listens to the conversation in real time.
But Sage is not a passive recorder. While it listens, it classifies what is being said into three categories: decisions, action items, and open questions. Each classification is timestamped and attributed to the speaker. When someone asks "Sage, what did we decide?", it responds aloud with a summary of decisions made so far.
When the meeting ends, Sage delivers a structured document within 60 seconds. Not a transcript dump — a clean, organised record with decisions at the top, action items with owners and due dates, and open questions that still need resolution.
Key capabilities:
Cara — Discovery & Deliverables
Status: Coming Soon
Cara picks up where Sage leaves off — specifically for discovery calls and client-facing meetings where requirements need to be extracted from messy conversations.
In a discovery call, people speak in half-formed ideas, implicit assumptions, and contradictory goals. Cara listens, extracts structure, and produces clean requirements documents, scopes, and client-ready deliverables. She identifies unknowns — the questions nobody thought to ask — and flags dependencies between workstreams.
Where Sage gives you a meeting summary, Cara gives you a Statement of Work. Where Sage captures what was said, Cara captures what was meant.
Key capabilities:
Dash — Agile Ceremonies
Status: Coming Soon
Dash is purpose-built for the recurring meetings that engineering teams run every day: standups, sprint planning, and retrospectives.
Standups are notorious for running long and losing structure. Dash facilitates by keeping time, prompting each participant for their update (what they did, what they plan to do, what is blocking them), and producing a structured follow-up for the team.
For retros, Dash captures what went well, what did not, and proposed improvements — then captures the agreed-upon changes so they do not get forgotten by the next sprint.
Key capabilities:
Why Three Agents, Not One
The temptation is to build one agent that does everything. We tried that. It does not work well for the same reason a general-purpose employee is less effective than a specialist.
Each meeting type has different goals, different outputs, and different definitions of success:
One model trying to serve all three contexts simultaneously produces mediocre results in each. Three specialised agents, each optimised for their domain, produce outputs that are actually useful.
The Compound Effect
One agent is useful. Three agents that cover the full meeting lifecycle are transformational. The compound effect comes from continuity: Sage captures the strategic decisions, Cara turns client conversations into actionable scopes, and Dash ensures the execution stays on track through daily ceremonies.
For a CTO, this means meetings stop being a black hole where decisions go to die. Every conversation produces a structured, actionable output — and the right output for the type of conversation that was had.