Peptide research is moving fast and the public record is messy — scattered across PubChem entries, trial registries, vendor PDFs, and forum lore. AmericanPeptide.com pulls that record into one place an AI agent can read, cite, and reason over — so researchers can see how a peptide is designed, synthesized, purified, and proven, with the sources attached.
Most peptide information online sits at one of two extremes: dense primary literature written for specialists, or marketing copy written to sell. The middle — accurate, sourced, plainly explained — barely exists. That gap is where bad decisions get made.
We're building the middle. A research-grade reference layered over the free public datasets, an AI agent that answers from those sources instead of inventing them, and tooling that turns a certificate of analysis or a reconstitution math problem from a guessing game into a checklist. The goal is simple: when someone asks a question about a peptide, the honest, cited answer should be the easiest one to find.
AmericanPeptide.com is the name on that effort because the domain should be an authority asset, not a storefront. We treat it that way.
A citation-backed research assistant for mechanisms, synthesis, and clinical evidence — with live grounding tools.
91 peptides with PubChem-enriched chemistry, sequences, identifiers, and editorial background.
Reconstitution and blend calculators, a COA decoder, a compound builder, and synthesis walkthroughs.
Everything here is for computational research and education. We publish mechanisms, chemistry, and evidence — never dosing protocols, diagnoses, or treatment advice.
The full 91-peptide catalog is free to read, query, and redistribute under CC BY 4.0 — with a markdown twin and MCP server so agents can cite it directly.
Agent answers and reference pages trace back to PubChem, UniProt, PubMed, and ClinicalTrials.gov. Claims carry citations; computational outputs are flagged as hypotheses.
The core tools — the Agent, the catalog, the calculators, the COA decoder — are usable without an account or a paywall. We earn trust before we ask for anything.
The Peptide Agent gets better when researchers tell us where it's right, where it's wrong, and what's missing. We run an open channel for exactly that — flag an answer, suggest a catalog addition, or compare notes with other researchers.
Peptide Agent feedback · invite-only
The Peptide Agent feedback loop. Drop a question the Agent got wrong, a study it should know about, or a peptide we're missing — corrections feed straight back into the catalog and the Agent's grounding.
@USPeptide
Release notes, new catalog entries, and notable peptide research as we publish it.
Corrections, data questions, partnership and supplier inquiries, or press — reach the team directly. We read everything and reply to research and correction notes first.
Research use only: AmericanPeptide.com is an AI-assisted computational research platform, not a medical device or clinical decision-support system. All outputs are computational hypotheses requiring independent experimental validation, and nothing here is medical advice or an offer for sale. American Peptide is independent and is not affiliated with, endorsed by, or representing the National Institutes of Health (NIH), the National Library of Medicine (NLM), the National Center for Biotechnology Information (NCBI), or any government agency; ClinicalTrials.gov, PubChem, PubMed, and UniProt are public databases we cite as sources.