Generates a structured verification checklist for scientific papers or research claims, covering retraction status, validity, and currentness. Useful before citing studies in published work.
How this skill is triggered — by the user, by Claude, or both
Slash command
/autopunk-media-skills:retracted-papers-checkerThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Produces a structured verification checklist for any scientific paper or research claim you plan to cite, covering every step a journalist should take to confirm the paper is valid, current, and not disputed or retracted.
Produces a structured verification checklist for any scientific paper or research claim you plan to cite, covering every step a journalist should take to confirm the paper is valid, current, and not disputed or retracted.
Required: The paper title, author names, and the journal it was published in. A publication year is strongly recommended. Optional: The DOI (digital object identifier — the unique code usually printed at the top of a paper), any specific claims from the paper you plan to cite, and the context in which you encountered the paper (press release, social media, a researcher's recommendation).
Important: the assistant cannot access live retraction databases, journal websites, or the internet. This skill gives you the right checklist and tells you exactly where to look — you carry out the checks yourself. This is by design: retraction databases update daily, and only a live search reflects the current status of a paper.
A numbered checklist of five to eight steps. Each step includes:
Followed by a short "Red flags to watch for" section (three to five bullet points) tailored to the paper's subject area.
Total length: 300–500 words. Plain language throughout — no assumption that the journalist has used academic databases before.
Paper title: "Dietary flavonoid intake and risk of type 2 diabetes: a prospective cohort study" Authors: M. Hartley, S. Okonkwo, L. Ferreira Journal: Journal of Nutritional Epidemiology Year: 2019 Claim I plan to cite: "Eating flavonoid-rich foods daily reduces the risk of developing type 2 diabetes by 18%."
Verification checklist for: "Dietary flavonoid intake and risk of type 2 diabetes" (Hartley et al., 2019)
Check for retraction — Retraction Watch Database Go to retractionwatch.com and search the paper title or lead author name. This is the most comprehensive public database of retracted papers across all fields. A result here means the paper has been formally retracted; note the stated reason (data fabrication, ethical violation, honest error). If the paper does not appear, that is a good sign — but retraction Watch does not index every journal, so this is not a final clearance.
Check the journal's official record — PubMed or the journal website directly Search PubMed (pubmed.ncbi.nlm.nih.gov) for the paper title. Look at the record for any tags reading "Retracted Publication," "Erratum," or "Expression of Concern." These appear even when the paper is not fully retracted but has known problems. If you cannot access the journal website directly, the PubMed record will show any formal notices the journal has issued.
Check for a correction or erratum A retraction is the most serious outcome, but corrections matter too — especially if the corrected figure changes the claim you plan to cite. On the journal's article page, look for a linked "Correction" or "Erratum" notice. For a 2019 paper, it is worth checking whether the 18% figure appeared in the original or was revised in a subsequent correction.
Check whether the finding has been replicated or challenged Search Google Scholar (scholar.google.com) for papers that cite this study, and look specifically for systematic reviews or meta-analyses on dietary flavonoids and diabetes published after 2019. Nutritional epidemiology is a field with known replication challenges — an isolated cohort study finding that has not been confirmed in subsequent research should be cited with appropriate caution.
Check the authors' conflict of interest declaration Return to the paper's full text (via PubMed or the journal) and locate the conflict of interest or funding disclosure section. Industry funding from food or supplement manufacturers does not invalidate a study, but it should be disclosed in your story if you quote the finding.
If you hit a paywall Try searching the paper title on Unpaywall (unpaywall.org) for a legal open-access version. Your organisation's library may also be able to retrieve it. Do not rely on the abstract alone if you plan to cite a specific figure — the methodology section is where the 18% figure would be explained and qualified.
Red flags to watch for in nutritional epidemiology studies:
npx claudepluginhub ur-grue/autopunk-media-skills --plugin autopunk-media-skillsVerifies academic citations using canonical sources (DOI, arXiv, CrossRef, Semantic Scholar). Provides principles for detecting fake citations and matching metadata.
Validates bibliographic references by searching online and comparing citations against authoritative sources. Detects incorrect author, title, year, URL, or identifiers.
Assesses trustworthiness of scientific publications by checking retractions, corrections, expressions of concern, and reproducibility failures via PubMed, bioRxiv, Consensus.