Use it from your own assistant
Everything on this site is also a tool your AI assistant can call: search the science as of a date, map a field, rank labs and companies, and run a company with every claim sourced.
Finding the address
Once connected, ask your assistant "What can Diaspore do?" to check it.
Ask it things like
Run a company yourself
Your assistant does the reading and Diaspore keeps it honest. It hands your assistant the company's annual report and trial records as they stood on your date. Your assistant sends back findings, each with an exact quote. Code drops any finding whose quote is not in the filing word for word, or whose number is not in its quote, and builds the report from what is left. Nothing is charged to anyone.
What it can do
| What it does | Tool name |
|---|---|
| Labs ranked for a topic as of a date, each with its matching papers and the claims a model extracted from their abstracts. | lens_search |
| Who had public evidence on a topic or target as of a date, by lab and institution, and how the field filled up. | field_map |
| The ranked list of young companies for a stage, with reasons and the ranking's past record. | companies_to_watch |
| The ranked list of labs as of the latest data date, with reasons and the ranking's past record. | labs_to_watch |
| Start a deal run your assistant reads: the dated set of dated source files and the reading instructions. | deal_start |
| The deal runs already made. | deal_library |
Everything else it can do
| What it does | Tool name |
|---|---|
| What exists here: tools, counts, the science corpus, measured accuracy, and the rules. | about |
| What one lab's extracted last-author papers say as of a date. | lens_lab |
| The science of a company's founding labs, where the lab head is named on the company's filing. | lens_company |
| A brief built ahead of time: each claim checked against retrieved abstracts, with two exact quotes. | science_brief |
| Every benchmark task with the system against each baseline, intervals, and how the answer key is built. | bench_results |
| One stored end-to-end run on a company as of a date: report, record with sources, claim-check result. | deal_run |
| One file of a run's set of dated source files, in pieces. | deal_file |
| Submit findings; code keeps only those whose quote and numbers match the source. | deal_findings |
| Submit the case for and against; returns the report built from the checked record. | deal_decide |
| The six stages, the agent for each job, and what it produces. | roster |
| Every finished test a lab or company was ranked in: its rank at the cutoff, and what happened after. | track_record |
| One lab as of a date: signals, team counts, lab-head-on-filing company links, recent papers, freshness. | get_lab |
| One company as of a date: financings from SEC financing filings, events, origin labs, signals. | get_company |
| Every finished test with its metrics and every baseline, including baselines that were not run and why. | hindcasts |
| The top of one finished test's ranking with the outcomes that followed. | hindcast_ranking |
| How many labs and companies sit on each rung from lab to Series B. | pipeline |
| Find a lab by lab head name or a company by name. | find |
Every tool follows the rules of the pages: results are dated, sources are cited, only heads of labs are named, and a company appears beside a lab only when the lab head is on its SEC filing. A tool that cannot meet a rule refuses instead of guessing.