Chat
The chat panel is the fastest way to get answers out of a drawing set. Ask anything in plain language; the agent reads the set, answers with citations, and can run real work for you: checks, comparisons, reports, filled-out forms.

Asking questions
Type a question the way you'd ask a colleague:
"Which sheets show the electrical room?"
"What's the fire rating on the corridor walls in Area B?"
"Does the door schedule match the doors tagged on the plans?"
Every answer comes with citations: click one and the viewer jumps to the exact spot on the sheet, so you can verify the answer yourself in one click.
You can also:
- Attach files in a project's chat (images, spreadsheets, PDFs, Word documents, CSVs, text, markdown, logs, JSON and YAML) for the agent to analyze alongside the set - up to 12 MB of files per message, as many files as fit. Larger files: upload them to the project's Files and ask about them there. See Fill in any form. The chat on the home screen can't take attachments - see the home chat.
- @-mention a document, a workflow or a resource: type
@in the chat box and pick from the list - Workflows, Documents, or one of your Resources (a file, or a whole folder, listed under its folder's name) - so the agent knows exactly what you mean. - Snip part of the drawing: with a document open, click the snip button in the chat box and drag over a region of the viewer to send it with your question.
- Ask about an issue: the Ask Structured button on an issue card starts a chat about that finding.
- Dictate instead of typing: where dictation is available, click the microphone (Dictate) and speak. Your words land in the chat box for you to check and edit - dictation never sends on its own, and it stops after 5 minutes.
- Stop a reply with Stop generating. The chat tells you what it had already done before you stopped it.
- Open project memory from the brain icon (Project memory) at the top of the chat.
- Go full screen (Cmd/Ctrl+Shift+F) to give the whole app your screen when you need room for a long thread.
Where the work actually happens
Each chat thread gets its own workspace - an isolated environment the agent works in for as long as the thread lives. That has three consequences worth knowing:
- Long jobs keep going. A multi-step task doesn't have to finish inside one reply. The agent works, reports back, and the workspace is still there for the follow-up.
- Follow-ups pick up where you left off. Step away, come back and ask "now do the same for the mechanical set" - the thread resumes with everything it already worked out, rather than starting cold. After a long gap - more than about six hours - the project opens a fresh composer instead, and the old thread is one click away in the tab strip or the history.
- It can produce real files. Reports, spreadsheets, Word documents, PDFs, filled-in forms and boards are written as actual files and saved to the project's Generated Outputs folder, as generated outputs.
You don't manage any of this. It's the reason the chat can be given a job rather than just a question.
Plan, explore, execute
For bigger jobs, the chat doesn't just answer, it works. Give it a multi-step task and it plans the steps, explores the drawings, and executes in one thread:
"Find every fire-rated wall, list the doors on them, and flag any missing a rating."
You can see what it's doing as it works, and a long step tells you how long it has been running. It cites what it finds. A reply that takes 10 minutes or more is emailed to you when it finishes if you've stepped away, and you can ask it to email you when this reply is done.
Anything that changes the project - an issue's status, an edit, a merge, a rename, an undo - is shown to you first on an approval card, and nothing happens until you approve it.
What it can do on your behalf:
| Ask it to | And it will |
|---|---|
| Search the set | Read pages, search text across every sheet, and pull up what it finds |
| Run checks | Dispatch checks from your library across a document or the whole project, or read and run the project's review plan |
| Run a code review | Review the set against the building codes selected for the project |
| Write a check | Turn a description or a pasted checklist into runnable checks - see Custom checks |
| Refine a check | Propose an edit to a check's prompt when its findings aren't landing right |
| Flag a problem | Raise an issue with a citation, straight into Issues & Comments |
| Pull existing findings | Read the project's open issues and reason over them, or find an issue by its number ("what's #14?") |
| Resolve, waive or reopen issues | Change the status of one issue or a whole batch ("waive every info-level issue on the M sheets"), after showing you which ones on an approval card |
| Assign issues | Assign or unassign people, one issue at a time or in a batch by sheet, discipline, severity or status. A batch adds people and never replaces who is already assigned |
| Edit an issue | Change an issue's title, description, recommended solution or severity |
| Merge duplicates | Fold duplicate issues into one. The copies are kept under Found by and can be split back out, and the kept issue takes the highest severity of the group |
| Clean up issues | Move an issue (its callout, its cloud, or where it's pinned), restyle it, export issues, or delete findings you don't want and undo that delete (deleting is for project owners only) |
| Recall earlier work | Search your previous threads on this project rather than re-deriving an answer |
| Set up a view | Open a document or line up an overlay for you to look at, including turning a sheet 90 or 180 degrees so it lines up |
| Measure | Measure a distance on the sheet at the drawing's scale, and draw the measured line |
| Inspect a model | Answer questions about an IFC model's elements and properties |
| Reprocess a document | Restart processing on a document that failed, or re-read one when you ask it to |
| Organize files | Move project files into folders or remove them. Every move and removal is listed in the project's History tab, where you - or the chat - can undo it |
| Tidy the project | Rename the project, or set its thumbnail from a sheet |
| Merge projects | Merge a project into a Bluebeam-connected project, so its files and issues live in one place (organization owners, owners of both projects, or team leads) |
| Work with RFIs | Draft a reply to an RFI you paste or upload, citing the sheets and spec sections it checked, and save it with the RFI in the project's RFIs tool (saving is for owners and editors). On a project linked to Procore, it also turns issues into Procore RFIs |
| Email you a file | Email you a file from the project, or a run's report, so you have it in your inbox. It only ever emails you, never anyone else |
| Email you when it's done | Email you when the runs going now finish, or every time from now on |
| Change a markup | Move or reword a callout that's already on the sheet, including issue clouds and their connected text - not just draw new ones |
| Draw a diagram | Make a board - a process, a sequence, a set of relationships - or change one you already have |
| Produce a deliverable | Write a report, build a table, fill in a form, generate a spreadsheet |
If it can't do something, it says so rather than guessing.
Changes to issues, files and the project itself are for the project's owners and editors. If you're a viewer on a project, the chat can read, search and answer, but it won't make changes to the project for you. If your organization role is view-only, the whole chat is read-only, and a report it writes comes to you as a download instead of being saved to the project.
Memory
Tell the chat something once, and it remembers - in whichever of four places fits what you told it:
- Your memory - how you personally like to work (units, tone, what you focus on). It is yours: only you can add to it or change it, from chat or from the Memory tab of My Dashboard.
- Project memory - a fact about this project's drawings or work. Everyone on the project benefits from it. Project owners and editors can add facts; viewers can read them but not add. You can remove a fact you added yourself; changing or removing someone else's is for owners and editors.
- Team memory - a house rule for one team, saved only when you say "for the team" or name the team. Members of the organization who can open that team see it. The team lead (or a lead of a team above it, or the organization owner) adds it directly; anyone else on the team can suggest one, which is sent to the lead to approve.
- Organization memory - a house rule for the whole organization, saved only when you explicitly ask for organization or company memory. Every member of the organization sees it (external collaborators don't), but only the owner can add to it directly - a member's addition is sent to the owner to approve.
You can remove a team or organization fact you added directly. One that went in through an approved suggestion is the lead's or owner's to change or remove, since they approved it for everyone. You'll get a notification when your suggestion is approved or declined.
If your organization role is view-only, you can read every memory you have access to, but you can't add, change or remove anything.
If you're an external collaborator (someone outside the organization), you have your own memory and the chat reads the project's memory, but you can't add to project memory, and team and organization memory are never used in your chats.
Chat picks the right one from what you say - "remember I prefer metric" goes to your own memory, "remember E-101 is electrical" goes to the project. Ask for organization or team memory by name when that is what you mean.
To see what is saved, open the Memory tab on a team's page for that team's memory. Your own memory and the organization's are on the Memory tab of My Dashboard, side by side. Everyone can read the organization's; owners (and anyone your organization's Permissions settings allow) edit it, and everyone else can suggest a change.
Every saved fact says who saved it and when. A fact the assistant saved from a chat reads "Saved by the assistant in Jamie's chat", with a link to the drawing that chat was open on when there was one; the person whose chat it was also gets a link back to the chat. If the assistant saved something wrong to project or organization memory, press Revert on it to take it out in one click (Undo on the notice puts it back). The person whose chat saved it can always revert it; otherwise it is the people who can change that memory, which an owner can widen in Org Settings > Permissions (under Memory).
Project memory
Project memory holds what you tell it as individual facts, in six kinds:
| Kind | What it holds | Example |
|---|---|---|
| Correction | Something the extraction got wrong | "E-101 is electrical, not civil" |
| Preference | How to work this project | "Always metric", "ignore the /Cut Sheets folder" |
| Context | A fact that isn't in the documents | "The east wing is already built and barricaded" |
| Decision | A human ruling, or how documents relate | "The 2026-04 set supersedes the March one" |
| Glossary | A project-specific term | "'WH' means water heater on this job, not wall hydrant" |
| Caveat | A data-quality note about the set | "Sheets A-201 to A-210 are scanned, not vector" |
These are facts you supply, never the agent's own conclusions - its findings live in Issues instead. That separation is deliberate: it means a fresh review is primed by your team's decisions and never by a stale verdict from a previous conversation.
How to use it:
- Save something: just say it. "Remember that we always review to the 2021 IBC on this job." A confirmation names which memory it went to - "Saved to this project's memory", say.
- See what it knows: ask, or open the brain icon (Project memory) at the top of the chat. "What do you remember about this project?"
- Take something back: say so. "Forget the metric preference, this job is imperial." The old fact is retracted, not deleted, so the record of what changed survives.
Everything active is loaded automatically at the start of every conversation, so you never have to re-brief a new thread. Checks read it too - a check run uses your recorded facts when judging what applies, which is why "this segment is built and barricaded" stops a code check from re-flagging it on every run.
Working across several threads
You're not limited to one conversation per project. Start a new thread for a separate line of work and switch between them - useful when a long job is running and you want to ask a quick question without disturbing it.
Each thread is a tab across the top of the chat. Click + (New chat) to start one. Double-click a tab to rename it, and close it with its X. Earlier conversations are in the chat list (All chats in project), which you can search. A teammate's thread opens read-only.
There is no archive. To remove a conversation, open the chat list (All chats in project) and click the trash icon (Delete chat) on its row; you can delete only your own chats. Closing a tab keeps the conversation in the chat list.
A long thread carries a lot of context, and past a point the agent starts losing track of earlier steps. The platform tells you when a thread is getting heavy ("This chat is carrying a lot of context") or near its limit, and a fresh one would serve you better - your runs and results are saved either way. Press Continue in a new chat: a new thread opens with "Continuing from my chat ..." ready in the box and the old thread's attachments carried across. Add what you want next and send; the agent looks up what it needs from the earlier thread.
The heavy-thread notice in a project chat, with Continue in a new chat
The home chat
There's also a chat on the home screen, outside any project, for questions that span projects ("which of my jobs have open fire-rating issues?") or for starting work before you've decided which project it belongs to. When it moves your work into a project, the project chat opens with "From your home chat. Press send to start." waiting for you.
The home chat is text only: it can't take attachments. To work with a file - a form to fill in, or a checklist PDF you want turned into a checklist - open the project and attach the file in that project's chat, or upload it to the project. Dropping files on the home screen, or Create project, starts a new project rather than sending the file to the chat.
Repeatable work
If the chat does something you'll want again, ask it to save the workflow so you can re-run it. See Workflows for saved multi-step reviews, and Custom checks for turning a standard into a runnable check.
When the chat recommends a tool you haven't used, it offers a short "Here's how" video inline. The full clip library is in the Help menu → User Guide.