Skip to main content
These docs ship with an AI assistant. If you’re reading this in a browser, you’ll see a chat button in the bottom-right of every page. It’s a search layer on top of the documentation — ask a question in plain language and it answers using only the pages on this site.

​
How it works

The assistant runs a retrieval-augmented search: when you ask a question, it finds the most relevant passages across the docs, then generates an answer grounded in those passages. It cites the pages it used so you can verify.
  • It only reads the docs on this site. It doesn’t know the current state of your workspace, it can’t see your customers, and it can’t take actions on your behalf.
  • Every answer is generated fresh from the docs. If the docs are updated, the assistant’s answers reflect the update as soon as the site is re-indexed.
  • Thumbs up / thumbs down on each answer is how you tell the team what’s working and what isn’t.

​
What it’s good at

  • “Where do I do X?” — Mapping a task to the right module (for example, “how do I mark a milestone complete”).
  • “What is X?” — Explaining objects and concepts like initiatives, stakeholders, the Customer Success Plan.
  • “How do I set up Y?” — Walking through the flows we’ve documented (inviting a teammate, creating a project, generating an MCP token).
  • Lifecycle framing — Helping you understand where a feature sits in Discover → Qualify → Plan → Engage → Deliver → Expand.

​
What it can’t do

  • Answer questions about your live data. It can’t tell you who your stakeholders are, what your renewal rate is, or whether a deliverable is late. Use the app for that.
  • Take actions. It can point you to the page and the button. It won’t click the button for you.
  • Know about features that aren’t documented yet. If a capability hasn’t been written up here, the assistant won’t surface it.

​
When the assistant says “I don’t know”

The honest answer is useful signal — it means there’s a gap in the docs, not necessarily in the product. Two things to do:
  1. Press the thumbs down on the answer. That flags the query for the team to review.
  2. Ask again with different words. Sometimes the phrasing didn’t match. “Invite a customer” and “add an external user to the workspace” might retrieve different pages.
If you’ve tried both and still nothing useful surfaces, check the most likely module page directly — the sidebar on the left groups pages exactly as the app does.

​
The AI assistant inside the app

Separately, Bluplai itself has an in-app AI assistant at app.bluplai.com. That assistant has access to your workspace data — your customers, projects, stakeholders — and can help you draft updates, summarize an account, or suggest next steps. It’s a different tool from the docs assistant described on this page. The in-app assistant is built to work two ways:
  • Page-scoped — open it on a specific customer, project, or module and it reasons over that surface. “Give me a quick summary of this customer,” “What’s at risk on this initiative,” “Who are the champions here.”
  • Workspace-scoped — open it from Home and it reasons across every account you have access to. “What’s the state of the portfolio this week,” “Which customers haven’t had an interview in 60 days,” “Where is adoption slipping.”
It can also populate modules from raw input — paste or upload a meeting transcript, a discovery PDF, or a set of notes, and the assistant extracts the relevant entries (goals, challenges, tech stack items, actions) into the right modules instead of you re-keying them. The exact surfaces and upload points are rolling out per module — check the assistant panel on each module page for the current coverage.

Open the app

Sign in at app.bluplai.com to use the in-app AI assistant on your live workspace.

MCP access

Build your own agents against the workspace with an MCP token.