DeepXiv · Drugs (drugs)

~257k US drug labels (openFDA SPL, latest version per set_id; Synced daily; exact counts are in corpus (live)). Indications, contraindications, interacting drugs, boxed-warning gist and one-line summaries are model-extracted; route / product type / pharmacologic class come from the openfda mapping block.

Corpus: 256,993 documents, updated 2026-10.

Setup

pip install -U deepxiv-sdk      # deepxiv registers a token on first use

With a registered key (https://data.rag.ac.cn/register), search and read are fair use: 60/min, 5 at once, 5,000/day; on the auto-registered token they spend the deepxiv daily limit at the credits listed below. ask draws on the daily ask allowance (resets 00:00 UTC; a site-wide daily total can run out early at peak times), or 3 lifetime trial asks on the auto-registered token. On 429, back off for Retry-After seconds. Data goes to stdout; cost, quota and sources go to stderr. Exit 0 ok, 1 API error, 2 usage error. The general CLI guide is the deepxiv-cli skill.

Start here

deepxiv fa search drugs "GLP-1 weight loss" -F has_boxed_warning=true -k 5
deepxiv fa search drugs "anticoagulant" -F route=ORAL -F is_rx=true -k 5
deepxiv fa read drugs f5e548d0-cc79-4c34-a3f5-e20a5b8b6564 -l brief
deepxiv fa ask "Which drugs interact with warfarin according to their labels?" -d drugs

Free samples: ids f5e548d0-cc79-4c34-a3f5-e20a5b8b6564; queries semaglutide weight loss.

Search

deepxiv fa search drugs "query" [-F FILTER]... [-k N] [--offset N] [--mode M] [--head] [--json]

Modes: hybrid, dense, bm25. Up to 100 hits. Cost 2 (with --head: 3). An empty query is rejected; use "*" to list by filters only. Filters combine with AND; an undeclared field exits 2.

Advanced filters (see deepxiv fa spec drugs): generic_name, substance_names, pharm_class_moa, manufacturer, rxcui, unii, product_ndc, application_number.

Read

deepxiv fa read drugs ID [ID...] [-l brief|detail|full] [-e EXTRA] [-p k=v]... [-f json|md|html]

At -l full, -p section=NAME reads one section and -p characters=N the first N characters.

Facets

deepxiv fa facets drugs FIELD lists the legal values and counts, free. Use it instead of guessing a spelling. Fields: route, product_type, pharm_class_epc.

Ask

deepxiv fa ask "question" -d drugs [-e low|medium|high] [-p k=v]... [--no-stream] [--json]

Agentic Q&A over this domain: the model turns the question into filters plus keywords from the domain's own self-description, reads the hits and answers with [drugs:<set_id> <title>] citations; /stream is NDJSON with real token deltas

Always pass -d drugs: automatic domain picking is still early.

What this corpus will and will not tell you

• 67% (171,903) are product_type=UNMAPPED: those SPLs have no openfda mapping block, so generic name / route / pharmacologic class may be empty — use spl_name for the product name.
• Only the latest version per set_id is kept; no label history.
• Indications and contraindications are extracted highlights; go back to the label for prescribing decisions.
• Not medical advice.
• The index holds no full text: level=full is not available yet (the label sections live on NAS (openfda/label); the index carries only section names and token counts).

Citing

Cite records as [openFDA SPL f5e548d0…]. Say plainly when the corpus has nothing.

Generated by scripts/gen_fa_skills.py from deepxiv fa spec drugs --json on 2026-10-19. Field names and prices can change; deepxiv fa spec drugs is the live source.