Yes. If your rates live anywhere in WordPress (a hidden rate page, ACF fields on a services post, or postmeta), SleekAI pulls them into the system prompt as context. The bot quotes the same per-word or per-piece numbers your proposals do, every time. When you raise rates on the page, the next conversation reflects it without any redeploy. For ranged pricing, the bot is configured to quote the range and then ask the qualifying questions that narrow it.

Yes. Tag your samples by beat or industry using a custom taxonomy. SleekAI filters by tag, so a fintech inquiry surfaces fintech work and a healthcare inquiry surfaces healthcare work. For long sample libraries the OpenAI Files vector store handles retrieval, so the bot fetches the two or three most relevant clips rather than pasting every link into one wall of text.

Yes. The system prompt encodes which beats you take bylined credit on and which you ghostwrite. When a brand asks about ghostwriting on a topic you only byline, the bot says so directly. Confidential ghostwriting clients can be referenced anonymously ("a Series B fintech") so the bot can still describe relevant experience without naming anyone you signed an NDA with.

Yes. Encode your minimums in the prompt: per-word floor, beats outside your interest, content types you don't take (paid posts, anonymous opinion pieces). When a prospect clearly doesn't fit, the bot says so politely and points to a relevant blog post, newsletter, or referral. They leave with a useful next step instead of a wasted hour, and your calendar stays clean.

BYO API key means you pick: OpenAI, Anthropic, Google, or OpenRouter. For most writer portfolios a mid-tier model handles the qualifying conversation easily, since the bot mostly retrieves your samples and rates. If you do a lot of research-heavy explainer work and want the bot to summarise samples in chat, a stronger model is worth the cost. Token usage per conversation is logged, so the cost picture is visible from day one.

The bot hands off to your existing scheduler (Calendly, SavvyCal, Tidycal) when the brief is qualified. It drops the link with email and project type pre-filled so the editor doesn't repeat themselves. A Zapier or webhook trigger can push the chat transcript into Notion, Airtable, or your CRM so the scoping call has full context before the camera turns on.

Conversations log to your own WordPress database, not a third-party SaaS. The AI provider sees the chat text in flight (this is true of any LLM-backed chatbot), but logs and storage stay on infrastructure you control. For sensitive conversations, you can disable logging per bot or auto-purge logs on a schedule using a standard WP cron job.

Yes, Multibot supports that. Run a fintech bot on your fintech samples page, a healthcare bot on the healthcare page, and a generalist bot on the homepage that triages between them after the first qualifying question. Each bot keeps its own conversation logs tagged with the bot ID, so analysis of which beat converts best is a database query, not manual transcript review.