Skill

Trial evidence trace

Paste one clinical-trial ID; get the exact published papers that describe it — pulled from Amass's publication↔trial graph, with trust metadata and abstract-grounded relevance notes.

Download skillSee the prompt

New to Amass? Connect Amass to Claude first

7linked papers

2trials traced

You getcsv2 files

Skillclinical-evidenceclaude

rank	title	journal	publicationDate	citationCount	journalQualityJufo	isRetracted	relevance	pmid	amassId
1	Efficacy and Safety of NVX-CoV2373 in Adults in the United States and Mexico	The New England journal of medicine	2022-02-10	438	3	false	Pivotal PREVENT-19 primary results: vaccine efficacy 90.4% (95% CI 82.9-94.6) vs symptomatic COVID-19 in 29,582 US/Mexico adults; 100% vs moderate-to-severe.	34910859	AMBC_RxICqodFU781soflbEBVHyIqqs8
2	Safety, Immunogenicity, and Efficacy of the NVX-CoV2373 COVID-19 Vaccine in Adolescents: A Randomized Clinical Trial	JAMA network open	2023-04-03	48	1	false	Adolescent (12-17y) expansion: vaccine efficacy 79.5% (82.0% for the Delta variant); neutralizing-antibody response non-inferior to young adults.	37099299	AMBC_I71wQrdCRzWOrZESANJrUC1AgI
3	Safety, Immunogenicity and Efficacy of NVX-CoV2373 in Adolescents in PREVENT-19: A Randomized, Phase 3 Trial	medRxiv (preprint)	2022-09-21	11	0	false	Preprint of the adolescent-expansion results (same 79.5% efficacy); peer-reviewed version is the JAMA Network Open paper above.	36172135	AMBC_4ubf9dLCJ8h028EYbYOm0CkDXf0
4	Association Between SARS-CoV-2 Viral Load and COVID-19 Vaccination in 4 Phase 3 Trials	The Journal of infectious diseases	2024-12-16	8	2	false	Pooled 4-trial analysis: NVX-CoV2373 associated with a 2.78 log10 copies/mL reduction in SARS-CoV-2 viral load at diagnosis.	39225478	AMBC_FxpOGrrHe0APnKskuvMtce15PG8
5	Modifiers of COVID-19 vaccine efficacy: Results from four COVID-19 prevention network efficacy trials	Vaccine	2023-07-25	8	1	false	Cross-protocol analysis of four CoVPN trials: vaccine efficacy vs symptomatic COVID-19 had little heterogeneity across baseline host/exposure factors.	37385888	AMBC_LUkekxNXcnw17HKN50r791HIXsy

The prompt

Paste this into Claude

Clinical-trial ID (NCT or AMTC)NCT04611802

prompt

Use the Trial evidence trace skill.

Clinical-trial ID (NCT or AMTC): NCT04611802

Why not just ask the model?

Ask a plain model "which papers describe NCT04611802?" and it invents plausible citations and PMIDs, because it has no real trial→paper map to read from.

Plain model: fabricates titles, PMIDs, and journals that "sound like" the trial's literature, and conflates papers that merely cite COVID-vaccine work with the ones that actually report this trial.
Amass: the trial record carries a first-class referencesBiomedCore edge — the Amass IDs of the publications that describe it. Each fetched paper returns its verbatim title, journal, publicationDate, citationCount, journalQualityJufo, and isRetracted, with a relevance note quoted straight from the returned abstract. Nothing is guessed.

Why a skill, not a prompt?

This is a multi-step orchestration, not a one-shot lookup: one get_amass_trialcore_record reads the trial's referencesBiomedCore array, then a get_amass_biomedcore_record by-id fan-out runs once per referenced paper to pull its trust metadata and abstract, then a templated roll-up sorts by citation count and emits the CSV. Packaging it as a skill means the rate-limit-batched fan-out — and the metadata-only recovery for trials whose landmark papers overflow the per-call token budget — ships once and triggers with a single line, instead of re-pasting a brittle procedure every time.

Honest scope

The describing-paper count is the trial's referencesBiomedCore length (7 for PREVENT-19), not an exhaustive literature search — broaden coverage by also searching BioMedCore for the trial's acronym and unioning the results. Citation counts and trust metadata are verbatim from Amass and can lag the source databases by hours. Every relevance note is grounded in the paper's own abstract; where the abstract is silent it reads "not in abstract", and no trial-outcome string is paraphrased from a field the MCP does not expose (it does not surface primaryOutcomeMeasures).

Install the skill

Add the skill to Claude

Install every Amass skill with one command — npx skills add amass-technologies/public-skills — or download the SKILL.md from GitHub and add it to Claude (Settings → Capabilities → Skills, or drop it in your Claude Code skills directory). Then paste the prompt above to trigger it.

Download SKILL.md View on GitHub

Keep exploring

csv

12targets scored

kinase	kinase_family_chembl_leaf	small_molecule_tractability	protac_tractability
PLK1	Other protein kinase PLK family	Advanced Clinical	predictive-only
WEE1	Other protein kinase WEE family	Advanced Clinical	predictive-only
CDK1	CMGC protein kinase CDK5 subfamily	Advanced Clinical	predictive-only

researchSkill

Target prioritization matrix

A general, interview-driven target-discovery workbench: name the biology and the modality you care about, and get back a scored table of candidate targets — druggability, DepMap essentiality, genetic constraint, curated safety, and the real number of drugs already targeting each — assembled from GeneCore intelligence and DrugCore/TrialCore/BiomedCore/PatentCore cross-links.

csv

2approved, first-gen

One-line read: approved, but already evolving. KRAS G12C went from "undruggable" to two approved drugs, and the field has immediately pivoted to a sec…

competitive-intelligenceSkill

Biomedical Evidence Scout

Name a drug, target, mechanism, or indication; get a landscape briefing that maps the published literature AND the clinical pipeline together — every paper and trial carrying a real Amass ID, PMID, or NCT you can check.

docx

5Phase-3 trials

1. Pivotal Phase-3 program

AMASS EVIDENCE BRIEF

regulatoryPrompt

Tirzepatide evidence brief

Name a drug asset; get a citation-auditable Word evidence brief — pivotal Phase-3 trials, grounded efficacy and safety findings, and the trial↔paper links — every figure verbatim from Amass.

Run it yourself

Connect Amass to Claude and paste the prompt.

No build, no deploy — the connector takes about a minute, then this workflow runs on your own inputs.

Connect Amass to Claude Browse all showcases