Persona Agents
AI that knows your role, not just your building
Four agents, four lenses on the same data. The asset manager sees lease risk. The facility manager sees equipment failures. The security director sees after-hours anomalies. Each agent speaks the language of the person using it.
Four role-tuned agents
Plain-English answers
Built on one data layer
Four agents. One data layer.
CRE Agent
Built for asset managers, portfolio leads, and investment teams. Tracks occupancy trends vs. lease expirations, flags sublease opportunities when utilization drops below threshold, and generates portfolio performance reports.
Sample Question
"Which buildings have occupancy below 60% with leases expiring in the next 18 months?"
- Occupancy vs. lease timeline
- Sublease opportunity flags
- Portfolio NOI impact
- Tenant retention risk score
Who it's for
Asset managers, portfolio leads, and investment teams.
Outcome
Maximize net operating income & minimize tenant churn.
Workplace Agent
Built for heads of workplace, experience managers, and hybrid work leads. Monitors desk utilization, meeting room demand, and employee flow patterns. Compares buildings and identifies floor consolidation opportunities.
Sample Question
"What's our average Tuesday utilization across the Chicago offices?"
- Desk utilization by day/floor
- Meeting room no-show rate
- Peak attendance patterns
- Floor consolidation candidates
Who it's for
Heads of workplace, experience managers, and hybrid leads.
Outcome
Optimal footprint utilization and frictionless employee flow.
Facility Management Agent
Built for chief engineers, maintenance directors, and FM operations. Correlates equipment runtime with energy consumption and maintenance history. Predicts failures and prioritizes condition-based PM schedules.
Sample Question
"Which rooftop units in my portfolio have the highest probability of failure in the next 90 days?"
- Equipment failure probability
- PM compliance rate
- Energy consumption anomalies
- Asset replacement timeline
Who it's for
Chief engineers, maintenance directors, and FM teams.
Outcome
Drastically reduced downtime and smarter energy usage.
Security Agent
Built for CSOs, physical security directors, and GSOC teams. Monitors after-hours access, correlates badge data with HR termination records, flags propped doors, and generates proactive threat detection reports.
Sample Question
"Show me all after-hours badge events at 660 Fifth Avenue in the last 30 days, cross-referenced with the current Workday roster."
- After-hours entry count
- Credential/termination mismatches
- Forced door events
- Insider threat correlation score
Who it's for
CSOs, physical security directors, and GSOC teams.
Outcome
Unified physical threat detection and instant roster parity.
MCP
Ask your data from the chatbot you already use
Savvy runs an MCP server, so Claude, ChatGPT, or any MCP-compatible assistant can query your building data directly. No separate Savvy login, no new app to open.
Get started with
Skip the extra login
Authenticate once through OAuth. After that, your assistant pulls live Savvy data on your behalf.
Bring your own chatbot
Connect Claude, ChatGPT, or any MCP client to the same data layer that powers the Persona Agents.
Persona Agents questions
Four: a CRE Agent for asset managers and portfolio leads, a Workplace Agent for workplace and experience teams, a Facility Management Agent for chief engineers and maintenance directors, and a Security Agent for CSOs and GSOC teams. Each is tuned to the questions that role actually asks.
Plain English, grounded in your building's real data. Ask about occupancy versus lease expirations, desk utilization across offices, rooftop units at risk of failure, or after-hours badge activity, and the agent answers using your unified data, not a generic model.
The last 7 days, unless you ask for something else. Ask about a specific building, date range, or portfolio and the agent resolves the window and tells you what it used.
Yes. Persona Agents and Dynamic Dashboards both read from the same unified data model that Savvy Connect builds, so an answer from the Security Agent and a number on the Access Analytics dashboard always agree.
The four agents are built around the roles we see most often across CRE and workplace teams. If your organization needs a different lens on the data, talk to us about what an enterprise deployment can support.