How Adam Smith Uses OpenClaw Agents for CRE Prospecting
Built by Adam Smith · · Source: Substack
Result: Topher Stephenson calls the printed daily briefing a "huge time saver." The source gives no numbers for deals, calls or hours saved.
What they built
Adam Smith is a student housing broker who built a team of OpenClaw agents to support their prospecting. Topher Stephenson wrote the setup up in the March 2026 edition of the ChatCRE newsletter, after Adam showed it in a live CRE AI Studio session.
According to Stephenson, Adam's first OpenClaw automation was a personal daily briefing. It pulls emails and internal messages, tags them by urgency and sends the result to a printer, so a paper brief is waiting at the desk each morning.
How it works
Stephenson describes five agents, each with a defined role:
- Atlas, the chief of staff. It runs on Claude Opus and stays on around the clock. Stephenson says it was trained on Adam's business plan and ChatGPT history, fills in missing data and keeps the other agents pointed at the brokerage's goals.
- Research agent. It scrapes the web daily and adds enrollment trends, population shifts, acceptance rates and sales comps to the student housing property database. It also works through hundreds of Google alerts and passes along only the relevant ones.
- News scraper. It watches LinkedIn and X and filters AI news down to what matters for the business that day.
- Olympus, the database agent. It fills gaps in Adam's Excel data, such as year built, ownership history and decision makers' LinkedIn profiles. That data feeds a CRM that ranks who to call each morning, flags likely seller signals and writes a brief for opening the call.
- Recruiting agent. It finds recruiting targets, builds lists and keeps the recruiting pipeline up to date.
The source doesn't give setup details beyond these roles. It doesn't name the CRM, the hardware or the costs, and it doesn't describe agents sending outreach or any approval step.
The result
Stephenson calls the daily briefing "practical, tactical, and a huge time saver." He gives no figures for deals, calls or hours saved. His main lesson from Adam's setup is that "Specificity is critical with AI": agents with clear roles, projects and goals do much better work.
How to build something similar
- Start with one small automation you will use every day, like a morning summary of your inbox, before you add more agents.
- Give each agent one narrow job and write down what it owns, what data it reads and what it hands off.
- Get your property or contact data into one place first. Agents that fill gaps are only useful if there is a clear record to fill.
- Point the output at a decision you make daily, such as who to call first, so you can tell quickly whether the system is helping.
The original post
Want to build something like this?
Join the Shipping Skool community free. Paid members get live bootcamp calls every week.
Join Shipping Skool — Free