You can now save and reuse your best prompts, search a new set of specialist life science databases, get an instant explanation of anything in a chat answer, and bring the Anara agent and your library into other apps you use. Here is what is new.

Save and reuse your best prompts with Skills

If you find yourself repeatedly entering the same kind of request, you can now save it as a Skill. Type / in any chat to open your prompt library, pick a Skill, and Anara drops the prompt in for you. Create and manage your Skills from Settings, where each one opens in the full chat editor, so you can build in @ mentions, files, and sources exactly like you would in a normal message. If you are not sure where to start, Settings includes a library of more than 100 ready-made Skills you can add in one click, spanning the life sciences, clinical practice and academia, with sets for law, finance, engineering, and policy too.

Anara for life sciences

New connectors let Anara reach specialist scientific platforms directly, making it easier to work across the discovery process, from literature review to data analysis.

BioRender connects Anara to its extensive library of vetted scientific figures, icons, and templates. The Protein Data Bank (PDB) provides access to experimentally determined 3D structures of proteins, nucleic acids, and larger molecular complexes. ChEMBL gives Anara curated bioactivity data on drug-like molecules and their targets. Open Targets surfaces the evidence linking genes and targets to diseases, alongside known drug pipelines. Connect any of them from Settings, and every result is cited back to its source database.

Bring Anara to your AI apps with MCP

You can now connect Anara to other AI apps that support connectors (via MCP, the open standard for linking AI tools), so they can search and read your library, and even call on the Anara agent itself, right alongside their own tools. Set it up from Settings under Connectors. Access is read-only: connected apps can read your library but never change it, and a clear consent screen lets you approve exactly what they can see before anything is shared.

Explain anything in a chat answer

Hit a term or sentence in a reply that you want unpacked? Highlight it and click Explain. A small popover streams a short, context-aware explanation of just that selection, without adding a new question to your conversation.

Two new capabilities for the agent this release, plus a refresh of how chats are found. The agent can now run code and hand you the result as a real file: a chart, a Word document, a spreadsheet, a quick analysis. Questions about figures, complex diagrams, scatter plots, gel images, dense tables, get much better answers. And chats are easier to come back to, with pinning, Cmd-K title search, and a URL per chat you can bookmark or share.

The agent can build files for you now

Ask for a chart from a dataset in your library, a Word document with your findings, a quick analysis, a spreadsheet, a LaTeX export. Each comes back as a file attached to the message, available on every plan as of May 15.

Much sharper answers on complex figures and tables

The agent handles complex visuals from your library much better now. Ask about something like a UMAP plot, a flow cytometry scatter, a gel image, a Western blot, or a dense data table from a paper, and the answer pulls from the specific colors, layouts, axis values, and marks in the figure. Table extraction and OCR-handling are tighter too, so scanned and OCR'd PDFs hand back cleaner structure for tables, printed page numbers, and figure captions.

Making chats easier to find

Three changes that work together. You can now pin any chat to the top of the sidebar from its dropdown menu. Cmd-K search now searches across all your chat titles, surfacing pinned chats by default and filtering live as you type. And every chat has its own URL: open a chat and the address bar updates with a link to that chat, so you can bookmark it, share it in Slack, or deep-link straight back into a thread from a note.

A handful of new touches this week: folder-level instructions for chat, in-app help, clickable cards for files and folders the agent acts on, and one-click connector setup from the chat input. Plus the usual long list of polish and fixes throughout the product.

Folder instructions

Each folder now has an Instructions tab in its settings. Whatever you put there is passed to chat whenever that folder is the active source. It's a clean way to set persistent rules ("always reply in Spanish", "cite in APA", "summarize before answering") for a specific research project without retyping them every message.

Ask Anara how to use Anara

Chat now knows the product. Ask things like "how do I import from Zotero?", "what's the difference between agent chat and ask chat?", or "how do I share a folder?" and Anara pulls directly from our help center to answer. No more bouncing out to support.anara.com to find the right article.

Jump to files and folders the agent works with

When the agent creates, moves, or otherwise acts on a file or folder during a chat, a clickable reference card now appears at the end of that message. One click takes you straight to it. Useful when you ask the agent to organize a literature review into a new folder, or take notes on a paper, and want to jump to what it just did.

Connect a connector from chat

Type "Notion", "Zotero", "Dropbox", or any other supported connector into the chat input and a pill appears above the composer to connect it in one click. Useful when you realize mid-question that you wanted Anara to pull from a tool you haven't hooked up yet.

Since Anara 3.0 shipped two weeks ago, we've been heads down on polish. This update covers the work we didn't have room to mention in the 3.0 post: a rebuilt note editor, a simpler chat input, the Related tab being retired, and a long list of fixes.

A rebuilt note editor

Notes are now powered by a new editor built from the ground up to work alongside Anara. The biggest day-one win is performance: long notes (10,000+ words) open immediately and stay smooth as they grow. AI edits stream in without the typing issues you used to see on long documents.

A few other things changed along with it. Equations get a dedicated input popover where you type LaTeX and see the rendered formula update as you go. Math is now parsed only inside $$ ... $$, so a single $ in a sentence stops getting silently turned into broken LaTeX. Toggles, callouts, and tables of contents persist correctly when you reload a note. Pastes from Word strip the heavy styling so your note keeps its own look. And familiar shortcuts like Cmd+Shift+8 for bullet lists work again.

@web replaces @search papers

Chat input source settings used to have separate @web and @search papers toggles, which was confusing because most web searches already pull in academic results. We collapsed @search papers into @web. One toggle now controls whether chat reaches outside your library, and Anara picks the right backend (general web, academic databases, life sciences connectors) based on your question.

We removed the Related tab

The Related tab is gone from the right sidebar. The suggestions were noisy, and it duplicated functionality already covered by chat and the Agent. Help guide: The Related tab has been removed.

A few months ago we started experimenting with ways to direct AI agents to complete research tasks end-to-end. Instead of simply working along side us summarizing and editing, we wanted a system that could handle research work autonomously across the full literature synthesis workflow.

Today, we're introducing Anara 3.0, a major update to our model harness and research workspace. We rebuilt Anara from the ground up to handle long-running research tasks, reviewing up to 100,000 files at once and with superhuman accuracy.

Give it a goal like completing a literature review or mapping the competitive landscape for a drug class and Anara will search your library, the web and academic databases, synthesize what it finds to return a cited, verifiable document you can edit.

We're rolling out access today and through the rest of this week.

What's new in Anara 3.0

Agent sidebar

Agent now lives in a new right sidebar; you can access it by clicking the “Agent” button in the top right corner of the app. Describe what you want and Anara will get to work. The most effective prompts tell Anara what to look at and what you want back. But don’t worry too much about the perfect prompt because Anara will ask follow-up questions.

You'll see it working in the chat thread, communicating which files it's reading and tasks it's carrying out. Once complete, it produces a final response either in the chat window or if it creates something it will be open as a new tab in the workspace. You can expand each section to understand how it reached a specific conclusion.

The agent sidebar stays open as you move around the app, and your chats are saved in a new "Chats" section in the left sidebar. By default Agent draws context from your library, sources from the web and the files and folders you have in view. You can disable library and web as sources with the buttons at the bottom of the input.

97% more accurate than general purpose AI

We evaluated Anara 3.0 against six general-purpose AI platforms on a multi-document question-answering benchmark (MADQA). Each platform was tested under identical conditions: same documents, same questions, same scoring criteria.

Anara 3.0 answers more questions correctly than every other platform tested, including ChatGPT and NotebookLM.

Anara 3.0 scores 97% higher than the average of all other platforms tested (43%) at locating the exact page containing the answer.

500M+ papers indexed

We expanded the number of academic papers we have indexed from the open web now amounting to 500M+ across sources like arXiv and JSTOR. Life sciences sources are now built in as connectors @-mention PubMed, bioRxiv, medRxiv, or ClinicalTrials.gov directly in chat to search them natively.

A more flexible interface

The agent works across many files, folders and notes in a single session, and until now that meant a lot of jumping between pages. To solve this problem we introduced a tabs feature. You can now open any file or folder as a tab and drag and split to work on files in parallel. Agent will always have context of the tabs in view.

Ask chat

The old way of chatting with files and folders hasn't gone anywhere. We just call it “Ask” chat now. Use it to keep a separate, ongoing conversation scoped to that specific file or folder. You can return to these chats just as you have previously by clicking the dropdown at the top of the chat thread.

Where people and AI do research work together

This won't be the last time the interface for doing science changes. More powerful models will unlock new interaction patterns. We're looking forward to continue building, simplifying and transforming Anara to be the place where people and AI do research work together.

These features were crafted by us but driven by conversations with you. Don't hesitate to reach out with thoughts and comments. We read every piece of feedback we get.