The new Amass app is live. Ask in plain language, check every claim. Try it free

Skills & PromptsRetraction hotspot scan

Prompt

Retraction hotspot scan

Name a research topic; get a checkable, downloadable list of the retracted papers Amass indexes on it — ranked by citation reach, with strict no-reason guardrails.

See the prompt

New to Amass? Connect Amass to Claude first

10retracted papers
2topic scans
You getcsv2 files
Promptresearch-integrityclaude

The prompt

Paste this into Claude

Research topicstem cell therapy regenerative medicine
prompt
TOPIC: stem cell therapy regenerative medicine

You have the Amass MCP connected (BiomedCore + TrialCore). Run a RETRACTION HOTSPOT SCAN
on the TOPIC above, for a journal/publisher research-integrity screen.

1. Call search_amass_biomedcore_records with:
     query = the TOPIC verbatim (strip any punctuation/operators — Amass is bag-of-tokens,
             no AND/OR/quotes; pick dense, distinguishing terms)
     isRetracted = true
   This returns up to `limit` papers that Amass flags as RETRACTED on this topic — `limit` is
   1–50 and defaults to 10, which is what this scan uses.

2. Build a table of every returned record, sorted by citationCount DESCENDING, with columns:
     Rank | Title (verbatim, unmodified) | Journal | Year | Citations | JuFo tier | PMID | Amass ID
   Then output the same table as a downloadable CSV.

3. GUARDRAILS — this is a MEDIUM-stakes integrity workflow. Obey exactly:
   - Report only the field-level facts Amass returns: the `isRetracted = true` boolean,
     the unmodified title, journal, year, citationCount, journalQualityJufo, PMID, Amass ID.
   - Amass exposes NO retraction-reason field. Do NOT infer or state *why* a paper was
     retracted. Never write "fraud", "misconduct", "fabrication", "plagiarism", "error",
     or impute intent. The only claim you may make about a row is: "flagged
     isRetracted = true in Amass as of <today's date>."
   - The search returns at most `limit` results (10 here) and no total count. State explicitly
     that this is a "top-10 sample by relevance, NOT an exhaustive census of retractions on the
     topic."
   - SPARSE-TOPIC CAVEAT: if the topic is narrow or its tokens are uncommon, an
     isRetracted=true search can fall back to off-topic retractions that merely share a
     token. Before trusting the list, sanity-check that each returned title is actually on
     the topic; flag and drop any that are not, and warn the reader that sparse topics
     degrade this way. (Dense topics like the example stay on-topic.)

4. Close with a one-line verdict, e.g.: "Amass flags N retracted papers on '<topic>' in this
   top-10 sample (as of <date>); highest-cited: <title> (<citations> citations, PMID <pmid>)."

Why not just ask the model?

Ask a plain model "which papers on stem-cell therapy were retracted?" and it confidently invents plausible titles, journals, and PMIDs — retraction status is exactly the kind of fact a model fabricates.

  • Plain model: invented titles, journals and PMIDs; guessed (often defamatory) "reasons."
  • Amass: a server-side isRetracted=true filter over a real index — 10 genuinely retracted papers ranked by citation reach (the MSC-exosome review still cited 201× after retraction, PMID 34253242), each a checkable record, with no inferred reason.

Honest scope

A top-10 relevance sample, not an exhaustive census (the search returns no total count). Each row claims exactly one thing — flagged isRetracted=true in Amass as of the run date; Amass exposes no retraction-reason field, so nothing here states why a paper was retracted. Sparse topics can surface off-topic retractions that merely share a token — eyeball the titles before trusting the list.

More from the library

Keep exploring

xlsx
14trials watched
Short nameSource idRegistryCopy of
Acute, 50/75 mg vs placeboNCT04072354ClinicalTrials.gov
↳ EUCTR2019-000470-36-BGEU CTRNCT04072354
Acute, 75/100 mg vs placeboNCT04092686ClinicalTrials.gov
competitive-intelligenceSkill

Watchlist Monitor

Name the trials, papers, approvals or drugs you need to follow; get one Excel board that re-reads them through Amass at every check and tells you exactly what changed: status, dates, enrollment, results posted, label revisions, retractions, new links.

pdf+1
196trials in scope
Landscape Monitor — page 1
competitive-intelligenceSkill

Landscape Monitor

Name a field; get it mapped across trials, papers, patents, drugs, genes and approvals into a landscape page with charts, a workbook and a one-page briefing, every record screened with its reason. Ask for an update later to see what is new and what changed.

csv
12targets scored
kinasekinase_family_chembl_leafsmall_molecule_tractabilityprotac_tractability
PLK1Other protein kinase PLK familyAdvanced Clinicalpredictive-only
WEE1Other protein kinase WEE familyAdvanced Clinicalpredictive-only
CDK1CMGC protein kinase CDK5 subfamilyAdvanced Clinicalpredictive-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, 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.

Run it yourself

Connect Amass to Claude and paste the prompt.

The connector takes about a minute to add, then this workflow runs on your own inputs.

Data
No training on your data
Hosting
EU-hosted, enterprise
Isolation
Tenant-isolated by architecture
Certification
ISO 27001, SOC 2, NIS2: in motion