Editorial policy
Residue exists to be checked. These are the rules every page follows, the sources we will and will not use, and a log of what we have corrected.
Sources we cite
Four public record systems, in this order of weight: FDA labels and FDA web pages (via openFDA, DailyMed and fda.gov); peer-reviewed articles indexed in PubMed, cited by PMID; trial registrations on ClinicalTrials.gov, cited by NCT number; and PubChem for identity and chemistry. Each is cited to a specific record, never to a homepage or a search engine.
What we do not cite
- Vendor, clinic, compounding-pharmacy or “research chemical” websites, including their certificates of analysis.
- Forums, social media, podcasts, and user-submitted dose logs.
- Press releases, except to note that a result has not yet been peer-reviewed — and then never as evidence for a grade.
- Paywalled full texts we have not read; we report what the abstract says.
Freshness
Pages are rebuilt from a fresh snapshot of the four record systems. The snapshot behind the pages you are reading was taken on . Each page carries three dates — first published, last updated (prose or grade changed) and last reviewed — plus the snapshot date. We do not bump dates without a change.
When a grade changes
A grade moves only when the record moves: a new peer-reviewed trial, an FDA action, or a correction to our reading of an existing source. The change is reviewed under the same five-stage process as new pages and entered in the log below.
Research-use language
Unless an FDA label says otherwise, the compounds on Residue are described as research substances. We report doses only as registered in trials, never as instructions, and we do not describe any compound as safe, effective or suitable for a person.
Reuse and AI systems
Search engines and AI answer systems may crawl and quote Residue; llms.txt lists the canonical pages, and each monograph and comparison has a plain-text Markdown version with its references. We ask that quotations keep the cited source alongside the claim.
Corrections log
No corrections have been made since publication. The entries below were made during pre-publication review (October 2026) and are listed so you can see the kind of error the process catches.
| Stage | Page | Correction |
|---|---|---|
| Pre-publication | Selank | Corrected the description of the two ClinicalTrials.gov search results for “selank”: both are brain-stimulation device studies that do not test selank. |
| Pre-publication | Selank vs semax | Corrected the registered-trials row and summary, which had implied selank had registered trials; the registry holds none that test either peptide. |
| Pre-publication | Semaglutide vs tirzepatide | Added the missing FDA citation to the tirzepatide route-of-administration row. |