Amass App
Ask in plain language. Check every claim.
Amass works through your question across eight specialised data layers — literature, scholarly papers, trials, drugs, genes, patents, the regulatory record and real-time web sources. It shows you every search it ran, and every claim it makes carries a citation you can open.
Ask
Start with the question you actually have
Type it the way you'd ask a colleague. The composer searches across the data layers at once, and the scope control lists exactly which. Its second mode, Build, is the next stop. New task is one keystroke away.
Build
Plan the report before it's written
Switch the composer to Build and the same question comes back as a plan instead of an answer. Amass scopes the work, then lays out the report's sections in order — each with its own sub-topics, and room to add more — so the shape of the thing is yours to change before a word of it is written. Run plan starts the work.
Show the working
Every layer it searched, every query it ran
Amass works in the open. The trace lists each data layer it went to and the exact search it made there, in the order it made them. When an answer looks off, you can see where it looked — and where it never did. Each claim in the answer carries a numbered citation, and the thread suggests where to go next — the patent angle, the latest news, the clinical evidence, or a full plan.
Citations
Cited and read are two different lists
Every reference opens — a DOI, an NCT number, a patent number, a regulatory label. Amass keeps what it cited apart from the wider set it read without citing, so the evidence behind an answer is something you can weigh rather than something you have to assume. Filter the references by data layer, hover a citation for the record, open it for the abstract and identifiers.
Trial landscape
The trials it cited, laid out as a field
Once a thread cites three or more clinical trials, a Trial landscape row appears at the top of its References card. Open it and the cited trials sit as phase lanes on one time axis, the most advanced phase on top, each bar coloured by its status. A bar splits at today — solid for the time elapsed, hatched where the completion date is still an estimate — and a trial with no dates on record never goes on the axis at all. Every bar opens its record in the references pane beside the chart, the legend filters by status, and the field grows as follow-up questions cite more trials.
Derived, not generated: the landscape is drawn from the records the answer cited, with no model in between, so it cannot show a trial the answer did not. In a finished report it sits beneath the body, and it travels into the PDF export.
Projects
Keep a line of inquiry together
A project groups the tasks and reports that belong to one question — a target, an asset, a landscape — so the work builds instead of scattering. Open a project and you see its tasks, each with the reports it produced.
Work with your team
One workspace, shared evidence
Your workspace has a name, members, and roles — owner, admin, member — and inviting a colleague is an email address away. Tasks, projects and reports live in the workspace, so the trace and citations behind a conclusion are there for whoever picks it up next.
Business: up to 10 seats with a shared workspace and usage pool. Every plan includes MCP access.
Reports
The thing you walk away with
A finished report is a document — contents, sections, every claim numbered to a source rail you can filter by data layer — with the plan it was built from, the map of what it covered and the full source log one tab away, and the chat kept beside it.
The sample below is a real report as it leaves the app, exported as a PDF from the button the Export stop shows — 31 pages on TROP2 antibody-drug conjugates available to license, 192 sources, every claim numbered to one of them. Published under the Amass name so it can be shared; the cover's QR code leads back to this page.
Report allowance by plan: 1 on Free, 5 on Basic, unlimited on Standard (fair use), unlimited and pooled across seats on Business.
Export
Take it with you
A finished report leaves the app as a PDF — one button in the report's header, Export report as PDF — and a chat thread exports the same way. Results tables download as CSV or Markdown. The citations travel with the document: the sample report on the stop above is one of these exports, all 192 numbered sources of it.
One engine, eight layers
One question reaches all of them
The literature, the scholarly papers, the trials, the molecules, the genes, the patents, the regulatory record and the real-time web are searched in one pass and cross-linked, so a result in one layer leads into the others instead of dead-ending. Six of them sit behind the API and MCP.
40M+
BiomedCore
233M+
ScholarCore
16M+
PatentCore
PreviewFDA + EMA
RegulatoryCore
22K+
DrugCore
1.2M+
TrialCore
43K+
GeneCore
Live
WebCore
Everywhere you work
The same engine inside Claude, ChatGPT and Cursor
Amass runs as an MCP server, included on every plan. Ask from the tool you already have open and you get the same layers and the same citations back.
Get started
Put it on a question you have already answered by hand
One question is enough to tell whether it holds up. Start on the free plan — no card required.