How Bioenergetic Oracle turns a question into a cited answer

Corpus scope and boundaries

The Peat-first, multi-author corpus contains roughly 150,000 passages from more than 50 years of writing and recorded discussion. It includes Ray Peat, Georgi Dinkov, Danny Roddy, and related researchers. You can see the represented voices or browse automated concept syntheses.

The corpus reports a particular bioenergetic and pro-metabolic tradition. It does not represent scientific consensus. The public interface shows cited source passages, not complete source documents. Use the title and source metadata to look for the original.

Choose between the two approved research paths on the Ray Peat research use-cases page. If you remember a topic but not the title, use the source-finding workflow to find Ray Peat source passages from a question.

If you already have wording or a claim attributed to Ray Peat, check an attributed quote against retrieved passages without treating the search as an authenticity verdict.

What an inline citation means

An inline citation points to a passage that the generated answer used as support. Open it to inspect the wording, attribution, source metadata, and available context.

The interface can flag low-confidence evidence or a weak or loosely related citation. Those signals help inspection, but they do not catch every retrieval or interpretation problem.

Standard answers and deep research

Standard mode fits a focused topic, author, claim, or comparison. Deep research uses broader, multi-round retrieval to look for gaps before it produces a longer synthesis.

Standard

Uses 1 daily-question slot and prepares one retrieval and synthesis flow.

Deep research

Uses 5 daily-question slots and can run additional retrieval rounds for multi-part questions.

Deep research can inspect a wider evidence pool. It does not guarantee a more accurate answer. See access limits and deep-research cost.

Automation, review, and corrections

Answers are generated automatically and are not individually reviewed by a person. Public concept pages are also synthesized automatically. The operator reviews core project copy, including this page.

There is currently no public benchmark for retrieval accuracy or citation faithfulness. That missing evidence is a limitation, not a performance claim.

To report an accuracy or attribution problem, email the operator. You can also read about the project and correction process. Rights holders can request a correction or takedown.

Known limitations

Source visibility makes an answer easier to inspect. It does not remove the failure modes of retrieval or generated synthesis.

• •Retrieval can miss a relevant passage, especially when a question uses unfamiliar wording.
• •A retrieved passage can be weak or tangential even when the answer cites it.
• •The synthesis can misread a passage, flatten a caveat, or connect ideas the source did not connect.
• •Authors in the corpus can disagree. One cited passage may not represent the full range of views.
• •The corpus has a defined intellectual scope. It does not stand in for current external evidence or scientific consensus.

Privacy and responsible use

Review the privacy policy to see how questions and account data are handled. Avoid entering information you do not want processed by the service.

The Oracle is a research starting point, not medical advice. Understand the health-information limits before relying on an answer. For consequential decisions, inspect the original source, compare current external evidence, and consult a qualified professional.