Browser research workspace
A browser research workspace is what you get when page context, notes, and memory all live in the same place instead of three separate tools
TabMate lives in your browser side panel. You ask from the page in front of you. You pin excerpts to hold across tabs. You save memories that TabMate recalls by relevance on future asks. You keep separate projects in separate named workspaces so nothing bleeds together. When you come back the next day, the context is still there. This page explains what each piece does and how they work together.
What keeps failing without a dedicated workspace
The failure is not the tools. It is that none of them are built to hold a research thread together across tabs, sessions, and projects simultaneously.
You open ten tabs, ask something in a chatbot, get an answer, close the tab, and start over tomorrow with no memory of what you found.
You save a useful quote but it is sitting in a note app with no connection to the page it came from or the question it was answering.
You are ten minutes into a new research session before you remember what you were actually trying to find out.
The six constructs that make the workspace work
Each one solves a different part of the continuity problem. Together they make browser research something you can build on instead of restart every session.
Workspaces
Separate containers per project
A workspace holds everything related to one research thread: conversations, memories, pins, saved prompts. When you are doing competitor teardown and customer pain mining at the same time, they do not bleed into each other. You switch workspaces in the nav bar — each one opens with its own context intact.
Page-grounded asks
Ask from the page you are already reading
TabMate lives in the browser side panel. When you ask a question, it uses the current page as context. If you have selected text, it focuses more on that. The question stays tied to the source — not to a vague description of it you typed into a chat window.
Memories
What TabMate remembers between sessions
Memories are information pieces TabMate retrieves by relevance when you ask something. You save them from excerpts, responses, or pins. They come in five types: facts (expire 180 days), snippets (60 days), summaries (60 days), preferences, and instructions. Preferences and instructions are always active on every ask — they do not need to match the topic.
Pins
Temporary holds that travel with you across tabs
Pins are excerpts you want TabMate to keep in front of it for the current session. Unlike memories, which are fetched by relevance, pins go into every ask — no filtering. They stay with you as you move between tabs. When you are done, promote them to memories or clear them. Stored per device, not on the server.
Memory candidates
TabMate flags what is worth saving
After each response, TabMate quietly scans the conversation and surfaces anything it thinks you should save. These show up as memory candidates for you to review. You approve or discard. Nothing gets saved automatically.
Internal state
A running note TabMate keeps for itself
After every ask, TabMate analyzes what has been done and makes a running note to stay aligned with your goals. This state draws from the current page, your memories, pins, and any instructions. It resurfaces important details as memory candidates so things you might miss do not disappear.
How a real research session flows
From first open to picking back up the next day.
Open the extension, pick your persona
On first use, you select what kind of research you are doing — marketer, founder, student, educator, or general research. This gives you starter playbooks curated for that workflow. A temporary workspace is created and you can start immediately.
Enable page access, start asking
TabMate needs to read the current page to use it as context. You enable access per site — it persists so you are not asked again on the same domain. Once enabled, select text to bring it in as an excerpt, or ask directly from the full page.
Pin what you need to hold, save what you need to keep
While working through a page or a line of inquiry, pin the excerpts you need TabMate to hold for the next few turns. When something is worth keeping long-term — a key claim, a customer quote, a fact to reuse — save it as a memory with the right type.
Switch tabs without losing the thread
Move to the next source. Your pins travel with you. Your memories stay in the workspace. TabMate knows what workspace you are in and keeps all of it scoped there.
Come back tomorrow and continue
The workspace holds your conversations, memories, and prompts. TabMate picks back up with the context that was there when you left. No summary needed, no copying things back in — just continue.
Memory types — what to use when
Memories are not one thing. The type you choose changes how they get recalled and how long they last. Picking the wrong type is not catastrophic, but picking the right one makes retrieval much more useful.
| Type | Use when | Expiry |
|---|---|---|
| Fact | A specific detail you will want to reference: a date, a price, a claim, a decision. | Expires after 180 days if unused. |
| Snippet | Verbatim text that matters as-is: a quote, a URL, a product name, a competitor positioning line. | Expires after 60 days if unused. |
| Summary | A condensed version of a longer thing — a page, a document, an earlier conversation. | Expires after 60 days if unused. |
| Preference | How you want responses formatted or toned. Applied to every single ask automatically. | Never expires. |
| Instruction | A rule TabMate should always follow: "cite sources", "keep replies short". Always active. | Never expires. |
Pins vs memories — the practical split
This is the question that comes up most. The short version: pins are for right now, memories are for later. But the details matter.
Use pins when
- You need TabMate to hold something for the next few turns.
- You are moving between tabs and need a reference point active the whole time.
- You have not decided yet whether something is worth keeping permanently.
- You want it included in every ask, not just when it seems relevant.
Use memories when
- Something is worth keeping across sessions, not just this work block.
- You want TabMate to recall it by relevance, not load it into every ask.
- It is a preference or instruction that should always apply, everywhere.
- You are building a reusable knowledge base for recurring research.
Signs this is the right fit
TabMate is not built for passive browsing or one-off questions. It is built for people who do repeated, source-heavy research and keep losing the thread.
- You do browser research regularly and lose context between sessions every time.
- You are comparing multiple sources and need claims, notes, and follow-up questions to stay attached to the right page.
- You run recurring research jobs — competitor teardown, review mining, vendor evaluation — and want a structured starting point.
- You need project separation so competitor research does not mix with customer notes.
- You want the AI to get smarter about your work over time, not start fresh on every ask.
FAQ
What is a workspace in TabMate?
A workspace is a named container for one research project or thread. It holds conversations, memories, pins, and saved prompts. Different workspaces stay completely separate — no cross-contamination between projects.
What is the difference between a pin and a memory?
Pins are temporary and always included in every ask — TabMate holds them for the current session and sends them alongside every question regardless of relevance. Memories are retrieved by relevance — TabMate matches them to your current question and only pulls in what fits. Preferences and instructions are the exception: they are always active like pins, but stored as memories.
What is internal state?
After every ask, TabMate makes a running note for itself about what has been done, what the goals are, and what context matters. This helps it stay aligned across a long session and surfaces important details as memory candidates instead of letting them disappear.
How does TabMate use the page I am on?
TabMate reads the active tab when you ask a question. You need to enable page access per site once — after that it persists for that domain. When you select text, TabMate focuses more on that excerpt. Questions stay tied to the source page, not to a paraphrase of it.
What are starter playbooks?
When you first open TabMate and select your persona — marketer, founder, student, educator, or general researcher — TabMate sets up starter playbooks curated for that workflow. They are pre-built prompt and workspace structures for the most common tasks in your field.
Does it work for multi-session research?
Yes. Workspaces hold conversations, memories, and saved context persistently. When you come back the next day or next week, everything is still scoped to that workspace and TabMate picks up from where you left off.
Related pages
These research jobs overlap. If this page is close to what you need, one of these may be too.
How to synthesize online research without losing context
A deep-dive guide to going from raw browser research to finished output: capture disciplines, multi-source swipe files, cross-source synthesis asks, and persona-specific workflows.
Read: How to synthesize online research without losing context
Best Chrome extensions for academic research and students
A practical extension stack for student research: citation tools, tab control, and source-grounded continuity for assignment workflows.
Read: Best Chrome extensions for academic research and students
How to do competitor research with AI in your browser
A 7-step workflow for capturing pricing, claims, and review signals from live tabs — keeping source evidence attached across the session.
How to group tabs by project without slowing down Chrome
A strict six-step framework for project-based tab grouping that controls tab sprawl while preserving source context across sessions.
Read: How to group tabs by project without slowing down Chrome