Yes. SleekAI uses OpenAI Files as a vector store for large documents, so CC&Rs, rules and regulations, architectural guidelines, and bylaws can all be indexed. The bot retrieves the most relevant 5 to 20 passages for each question and answers from those, citing section numbers where helpful. Update a document and reindex; no full bot rebuild is required to reflect amendments.

Multibot. Each community gets its own bot, with its own document scope, greeting, dues figures, and pool hours. Display conditions match the bot to the community page or homeowner portal section, so an Oakridge homeowner only ever talks to the Oakridge bot. Reports stay scoped per community, which makes board reporting straightforward.

Yes. The bot collects category, unit, urgency, photos, preferred contractor (if any), and access notes, then forwards a structured ticket to the portfolio manager by webhook. Many companies pipe this directly into AppFolio, Buildium, or Vantaca. The homeowner gets a confirmation reference number in the chat, and the conversation log keeps a copy in WordPress for audit.

ARB submissions captured in the chat include unit, proposed change, materials, contractor, expected duration, and photos. The bot pre-screens against the published architectural guidelines so obvious-yes and obvious-no cases are flagged for the ARB chair with notes attached. The chair gets a structured payload rather than a free-form email, which speeds up board votes.

The bot is instructed never to provide legal advice on covenant disputes, fines, or governance matters. Instead it offers to escalate to the portfolio manager or, when appropriate, to the association's legal counsel. The same applies to fair housing questions, lien procedures, and any tax matter relating to special assessments. The instruction is enforced in the system prompt and reinforced by the guideline filter.

Logged-in board members can see additional data the public bot does not surface, such as financial summaries, reserve study figures, or vendor contract details, controlled by user role. A separate bot scope for board members is common, with display conditions targeting a 'board' user role or a private dashboard URL. The public homeowner bot stays strictly within homeowner-facing scope.

Most HOA management offices report that 40 to 60 percent of inbound calls are routine: dues, hours, rules, and where to find a form. A bot that answers those instantly with the right policy reference cuts call volume and lets the portfolio manager focus on real escalations. The conversation log shows which question categories convert to tickets, which informs future homeowner communications and document updates.

Yes. SleekAI uses your own API key, so OpenAI, Anthropic, Google, or OpenRouter terms apply directly to your account. Standard OpenAI API terms do not use data for training. Conversation logs stay in your WordPress install. For management companies subject to specific data agreements with their associations, this is the correct architecture, and OpenAI's enterprise terms strengthen privacy further if needed.