Gosset CLI

The Gosset CLI gives you fast, agent-friendly access to the Gosset database from your terminal (drugs, clinical trials, companies, deals, and news), plus the trial-success (PTRS) and timeline models.

It ships in the same package as the Python SDK: pip install gosset gives you both the gosset command and import gosset.

Why the CLI

• You type names, not IDs. Pass a target, disease, drug, company or modality by name: the CLI resolves it to the right Gosset id under the hood. --target PD-1, --disease "atopic dermatitis", --drug pembrolizumab all just work. (You can still pass a raw id or class-id if you have one.)
• JSON by default. Every command prints JSON to stdout: pipe it into jq, feed it to an agent, or add --table for a human-readable view.
• One command per entity. gosset drugs, gosset trials, gosset companies, gosset deals, gosset news, with consistent flags across all of them.

60-second tour

pip install gosset
gosset get-token                     # one-time auth (or set GOSSET_API_KEY)

# The TL1A landscape, most-advanced first
gosset drugs --target TL1A --table

# Every TROP2 ADC/therapeutic (332 of them), Approved → Ph1
gosset drugs --target TROP2 --limit 20 --table

# Semaglutide trials
gosset trials "semaglutide" --phase 3 --table

# Recent TL1A / IBD deals
gosset deals --target TL1A --table

# Feed results straight into an agent or jq
gosset drugs --target "KRAS G12C" --phase 3 | jq -r '.[].name'

Real output for gosset drugs --target TL1A --table:

name          latest_phase  status  developers
------------  ------------  ------  --------------------------------------------
tulisokibart  3             Active  Schering-Plough, Prometheus Biosciences
duvakitug     3             Active  Teva Pharmaceutical Industries, Sanofi
afimkibart    3             Active  Roche, Roivant Sciences, Pfizer, Telavant
FG-M701       2             Active  Beijing Mingzhi Zhishan Biotechnology, AbbVie
…             …             …       … 49 total

The mental model

You pass	The CLI does	Sent to the API
--target PD-1	resolves the name → target IDs	targets=<ids>
--target GT-11-01	recognizes a class-id	target_classes=GT-11-01
--target 5f3a… (24-hex)	recognizes an ObjectId	targets=5f3a…
--disease "atopic dermatitis"	classifies → disease class	disease_classes=<id>
drugs "pembrolizumab"	resolves the drug name	drug_names=<ids>

Add --debug to any command to see exactly what got resolved and sent.

A note on unresolved names

If a name can't be resolved (typo, not in the DB), the CLI returns an empty result and a clear message: it never silently drops the filter and dumps the whole database.

$ gosset drugs --target notarealtarget
No match: could not resolve 'notarealtarget' (target). Check spelling, or pass a Gosset id.
[]

Where to next

• Getting started: install, authenticate, output formats, global flags
• drugs · trials · companies · deals · news
• Predictions: ptrs and timeline
• Recipes: real-world workflows (landscapes, competitor scans, agent pipelines)