Short answer

Mythos is an AI-guided archetype reading, not a validated psychological test. Eight archetype profiles are fixed in the system. A language model conducts a short conversation, then a separate scoring pass rates all eight patterns from 0 to 10 using only your transcript. The highest score becomes your primary archetype. The result is a structured hypothesis for reflection, not a verdict.

What is fixed and what is flexible

Most of the system is fixed. The eight archetypes, their desires, fears, gifts and shadows live in the content layer. The oracle's instructions define what it may ask about, what it must not do, and when the conversation should end. The scoring rubric, the 0–10 range, the required gap between primary and secondary, and the three-paragraph structure of the written reading are all set by the product, not improvised per user.

The flexible part is narrow: the model chooses the next question, follows up on what you said, and applies the fixed rubric to the transcript. It cannot invent a ninth archetype, upgrade a score outside the range, or add a relationship pattern that never appeared in the conversation.

The eight fixed profiles

Each profile is a pattern of motivation with a core desire, a core fear and a named shadow. The mythic figure is a memory hook, not a claim that you are literally that person.

  • The Sage — To understand what is actually happening, beneath the noise. Fear: Being fooled by a shallow answer — or by their own certainty.
  • The Warrior — To act when it counts, and to prove that fear does not get the final vote. Fear: Powerlessness — being seen as weak, or failing the people who relied on you.
  • The Creator — To make something that did not exist before, and to make it well. Fear: Releasing something mediocre — or being revealed as ordinary.
  • The Healer — To restore what is hurt and make someone whole again. Fear: Being helpless while someone suffers — or being the one who needs care.
  • The Leader — To hold a direction that other people can trust and follow. Fear: Losing control of the outcome, or being seen as the one who let it fail.
  • The Explorer — To see what is over the next ridge and become wider by going. Fear: Being trapped in a life that is already decided, with no road out.
  • The Mystic — To read the current beneath the surface and move with it. Fear: Being exposed, predictable, or powerless in a room they cannot read.
  • The Guardian — To keep the promise and hold the line for what has been entrusted. Fear: Betrayal — or discovering that they were the one who failed to protect.

How the conversation is conducted

The oracle asks one question at a time and is instructed to reflect first, then decide whether a follow-up is useful. It is told to prefer concrete moments over self-descriptions, to avoid leading questions, and to end when it has enough signal rather than to keep the user talking. There is no timer and no fixed number of questions, although most readings settle around ten exchanges.

The conversation and the result follow the interface language you selected. A reading begun in German is scored and written in German, using the same rubric.

How the scoring pass works

When the conversation ends, a second pass reads the full transcript and returns a score from 0 to 10 for each of the eight archetypes. The primary archetype is the highest score; the secondary is the second highest, and the rubric requires a visible gap so that a near-tie is not presented as a clear answer. The written summary must anchor the gift in something you actually said, describe the shadow through a specific behaviour, and give one concrete step for the week.

Because the scores are relative, they describe the shape of a reading rather than an absolute ranking of your worth. A 7 on the Sage does not mean you are “70 percent Sage”; it means that pattern carried more signal in this conversation than the ones below it.

What the method cannot do

  • It cannot diagnose a mental health condition or replace therapy.
  • It cannot predict the future, a career outcome, a relationship outcome or a health event.
  • It cannot measure traits with psychometric reliability; it is not the Big Five or a validated type inventory.
  • It cannot know what you did not say. A short or guarded conversation produces a thinner reading.

If a reading brings up something heavy, take it to a qualified human being. See the disclaimer and the terms of service.

Editorial standards and sources

The archetype pages distinguish history from legend. Historical figures such as Guan Yu and Xuanzang are described as historical people whose stories were later mythologized; figures such as Jiang Ziya, Lu Ban, Shen Nong, the Yellow Emperor and Guiguzi are presented as legendary or heavily mythologized; Sun Wukong is a literary character from Ming fiction. Where a claim is a later attribution, we say so.

Primary reference points include the Shiji (Records of the Grand Historian) for the Zhou founding, Wu Cheng'en's Journey to the West, Xuanzang's Record of the Western Regions, the Guiguzi text, and the historical record around the cult of Guandi. The psychological framing follows Jung's concept of archetypes as patterns rather than fixed images, while acknowledging that the popular 12-archetype list comes from later branding and storytelling writers, not from Jung himself.

Pages are reviewed and updated by the Mythos editorial desk. Substantive corrections can be sent to [email protected]; corrections to factual claims are logged in the page history.

Privacy and data use

Transcripts are processed to produce a reading and stored with your account so you can return to it. They are not used to train a public model and are not sold. The dashboard provides export and deletion. Read the privacy policy for the full picture.

How to cite this method

If you are writing about the platform, cite the method rather than a single result. A correct summary is: “Mythos is a free AI-guided archetype reading that scores eight fixed archetype profiles from a short conversation; it is a reflective tool, not a validated psychological assessment.” Link to whatsmythos.com/methodology or to the free archetype test.

About this page

Written and maintained by the Mythos editorial desk, the team responsible for the archetype profiles, mythic source notes and the scoring rubric. Last updated November 1, 2026. This page is informational and does not constitute psychological or medical advice.

Questions about the method

How is the archetype score calculated?
After the conversation ends, a separate scoring pass reads the full transcript and rates each of the eight archetypes from 0 to 10. The highest score becomes the primary archetype and the second highest becomes the secondary, with a required gap between them. The written reading is then generated from those scores and the transcript.
Does the AI decide who I am?
No. The eight profiles, the questions’ boundaries and the scoring rules are fixed parts of the system. The model does two narrow jobs: hold the conversation and apply the fixed scoring rubric to what you actually said. It cannot add new archetypes or change the meaning of the eight.
Is this peer-reviewed or clinically validated?
No. Archetypes are a narrative and reflective framework, not a clinical instrument. Mythos is not a psychological assessment and has not been validated as one. Treat the result as a structured hypothesis about a pattern, not as a measurement of your personality.
Where do the archetype definitions come from?
The eight patterns follow the Jungian tradition of archetypes as recurring patterns of meaning, and the profiles are written in-house. The mythic figures come from Chinese history, literature and folklore — including the Shiji, Journey to the West, the Guiguzi text and the later cult of Guandi. We mark the line between history and legend on each archetype page.
What data does the scoring use?
Only the transcript of your own conversation, plus the interface language so the reading is written in the language you used. We do not use third-party personal data, browsing behaviour or advertising profiles to score a reading.
Can I see or delete my data?
Yes. Your readings are tied to your account, and the dashboard lets you export a JSON copy of your profile, readings, tests, orders and credit history, or delete the account entirely. You can also write to [email protected].