They are not rivals. Learning the Tarot is a free, self-paced series of 19 lessons whose only real requirement is a tarot deck; it teaches card meanings and spreads in order, illustrated with the Rider-Waite deck. Wenbu is a browser workspace that draws tarot cards, casts I Ching hexagrams, calculates bazi and ziwei charts, keeps local notes and exports JSON, currently free without registration, with separate daily limits for AI interpretation. If your goal is systematic study, follow the course. If your goal is to record one specific draw, preserve the raw result and export it, Wenbu is more direct. The efficient path is often both: the course supplies method, Wenbu supplies reproducible calculation and savable records, while random draws stay documented rather than treated as repeatable predictions.
Learning the Tarot · Joan Bunning official course ↗Wenbu · current free access ↗Wenbu · calculation conventions ↗
- The course answers how to learn; Wenbu answers how to calculate, record and export. They can run in parallel.
- Wenbu's tarot and hexagram results are random draws, so record the result instead of expecting a repeatable prediction.
- Free access and rate limits follow the current page; after AI quota runs out, charting, drawing, notes and export still work.
A combined workflow from question to record
- 01Set the practice goal
Decide whether this session is about learning card meanings or working one concrete question; the course suits the former, the tool the latter.
- 02Choose the draw
Draw one or three cards in Wenbu, or draw by hand following a course exercise; Wenbu's draw is random and does not repeat within one draw.
- 03Keep the raw result
Write the card names, orientations, keywords and your first reaction into a note, separating card prompts from real-world facts.
- 04Request interpretation if needed
Read it yourself, request an AI interpretation, or hand it to an external agent; AI quota is limited and interpretation is not validation.
- 05Export and revisit
Save the material with JSON export, compare it with actual experience later, and decide what to study next.
Learning the Tarot · Joan Bunning official course ↗Wenbu · current free access ↗Wenbu · calculation conventions ↗
What each one actually solves
Learning the Tarot is explicit about its shape: a self-paced series of 19 lessons that begins with basics and moves gradually into more detailed aspects of tarot, with exercises in each lesson to reinforce the ideas, plus a Cards section for looking up individual cards as you go. It is geared toward beginners, though the author notes experienced readers will find useful ideas and techniques too. The only real requirement is a tarot deck; the online version is illustrated with the Rider-Waite deck, and if you already own a different 78-card deck you may use it, because the principles of tarot work are much the same whatever deck you use. What the course supplies is sequence, practice and reference, not a system that draws cards for you or stores your records.
Wenbu sits closer to a workbench. Its currently free, no-registration features include bazi charting with a five-element composition chart, the twelve ziwei palaces, I Ching casting with three coins, 78-card tarot draws, local notes, JSON export and a knowledge handbook, and the MCP and CLI calculation interfaces are also open without payment. You do not have to finish a course before using it; you can draw or cast, keep the raw result, then decide whether to read it yourself, request an AI interpretation, or hand it to an external agent. The two solve different problems: the course trains your judgment, the tool preserves your material.
Learning the Tarot · Joan Bunning official course ↗Wenbu · current free access ↗Wenbu · calculation conventions ↗
Randomness, recalculation and the limits of interpretation
Wenbu separates results from interpretation. Bazi and ziwei charts are calculated from the input data, while tarot and online hexagram casting are drawn randomly by the program, and the page keeps the raw result. Tarot draws one or three cards with equal probability from 78 cards, without replacement within a single draw; when reversals are enabled, each card independently has a 50 percent chance of appearing reversed. The three-card positions are present, pull and next step, used to organize thinking rather than to assert a necessary order of events. These are calculation and presentation conventions, not claims about the future.
The key difference is reproducibility. A deterministic chart can be recalculated from the same inputs and rules, so when comparing two results you should first check the inputs and calculation conventions. A random draw cannot be reproduced from the same question, so the honest practice is to record what was drawn rather than claim it can be repeated. Wenbu also states plainly that reproducible calculation does not mean the interpretation has been scientifically validated, and that traditional meanings, AI inference and the real-world details you supply should be kept separate. Learning the Tarot likewise frames tarot as a tool for personal growth and insight, saying you do not need psychic powers, only the willingness to honor and develop your natural intuition. Neither side promises predictive accuracy.
Wenbu · calculation conventions ↗Learning the Tarot · Joan Bunning official course ↗
Quota, records and export: the practical trade-offs
Wenbu's free access and its quotas should be read separately. Tool interfaces are currently rate-limited by request IP to 60 requests per minute, and users on the same network may share that limit. Browser visitors and signed-in accounts get up to 5 tool-page AI interpretations and 12 agent conversation turns per day, resetting at midnight Shanghai time, and signing in merges that day's trial usage rather than granting fresh quota. Shared networks have an additional anti-abuse cap of 50 interpretations and 120 conversation turns per day, and all visitors share a daily site-wide budget of 1,000 model requests. After quota runs out, you can still chart, draw, cast, save notes and export results, and you can route calculation results to your own AI assistant through MCP.
On records, Wenbu's local notes and JSON export let you keep a draw together with the question, context and result, and when exporting for an agent you decide whether raw birth data is included; export does not automatically upload anything to an agent. Learning the Tarot's materials are reading-and-exercise oriented, and the course itself does not provide a draw log or export mechanism, so exercise results usually live in your own paper or notes. If your core need is a long-term, reviewable record, Wenbu fits better; if your core need is understanding built in teaching order, the course fits better. Used together, let the course shape what you ask and how you read, and let Wenbu preserve what was drawn.
Wenbu · current free access ↗Wenbu · agent integrations ↗Learning the Tarot · Joan Bunning official course ↗
Three reader scenarios and a short evaluation list
Scenario one: a complete beginner who wants to understand the 78 cards first. Work through the Learning the Tarot lessons and their exercises, and use Wenbu's 78-card draws and knowledge handbook for cross-reference, but do not lean on AI interpretation too early. Scenario two: someone with some background who wants a record of one concrete decision. Draw one or three cards in Wenbu, save a note and export JSON, writing the cards, keywords and your own observations together, then revisit later. Scenario three: someone who wants to plug the workflow into an existing agent. Use MCP so the host model calls the calculation tools; Wenbu returns structured results and calculation conventions, the host model does the interpreting, and data handling follows the host's rules.
For evaluation, ask four questions. First, do I need a learning method or saved material? Second, do I accept that random draws are not reproducible and will I record results instead? Third, does my usage fit inside the free access and rate limits? Fourth, do I need to export results or hand them to an external tool? Writing the answers down before choosing is more reliable than choosing a tool and then finding reasons. Whichever path you take, do not blend traditional meanings, AI inference and real-world observation into a single conclusion.
Learning the Tarot · Joan Bunning official course ↗Wenbu · current free access ↗Wenbu · agent integrations ↗Wenbu · calculation conventions ↗
| Dimension | Learning the Tarot | Wenbu |
|---|---|---|
| Core role | Free self-paced 19-lesson course with exercises and card reference | Browser workspace for draws, casting, charting, notes and export |
| Prerequisites | A tarot deck; the online version is illustrated with Rider-Waite | Currently free and usable without registration for main features |
| Randomness and recalculation | The course involves no programmatic draw; readers do the exercises themselves | Tarot and hexagram results are random draws with raw results kept; charts are recalculable from inputs and rules |
| Records and export | The course itself offers no draw log or export mechanism | Local notes and JSON export, with export contents chosen by the user |
| Quota and limits | Free online material; no quota mechanism described on this page | Tool interfaces limited to 60 requests per minute per IP; daily limits on AI interpretation and agent turns |
| Best for | Readers who want card meanings and spreads built in order | Readers who want a reviewable record of one specific draw |
Learning the Tarot · Joan Bunning official course ↗Wenbu · current free access ↗Wenbu · calculation conventions ↗Wenbu · agent integrations ↗
Hypothetical practice: logging a three-card draw (fictional example)
This is a hypothetical example used to illustrate the workflow. It involves no real personal data and is not a calculated birth chart or a real draw. Suppose a reader wants to practice the present, pull and next step three-card reading.
- Step one: write down one practice question on paper, noting that this is practice, not prediction.
- Step two: draw three cards in Wenbu and record the card names and orientations; if reversals are on, each card independently has a 50 percent chance of being reversed.
- Step three: write three observations yourself first, then read the keywords, and only then consider requesting an AI interpretation.
- Step four: put the question, card names, keywords, your observations and what actually happened that day into a local note.
- Step five: export JSON to save it, and a week later mark which prompts helped and which were only your own associations.
When you revisit, judge only whether the record is clear and the observation honest, not whether the cards were accurate. The value of the practice is accumulated comparable material, not a prediction scoreboard.
Wenbu · current free access ↗Wenbu · calculation conventions ↗Learning the Tarot · Joan Bunning official course ↗
Questions you may have
Should I start with the course or the tool?
It depends on the goal. To build card meanings and spreads systematically, start with the 19 Learning the Tarot lessons and their exercises. To keep a record of one specific draw, draw in Wenbu, write a note and export it. They can also run in parallel: the course supplies method, the tool supplies records.
Learning the Tarot · Joan Bunning official course ↗Wenbu · current free access ↗
Can Wenbu's tarot draw repeat the same set of cards?
Not reproducibly from the same question. Wenbu draws tarot randomly with equal probability from 78 cards, without replacement within one draw, so the honest practice is to keep the raw result rather than expect the same cards again. Only deterministic charts are recalculable from the same inputs and rules.
Wenbu · calculation conventions ↗Wenbu · current free access ↗
What can I still do after the free AI quota runs out?
According to the current page, after AI quota runs out you can still chart, draw, cast, save notes and export results, and you can route calculation results to your own AI assistant through MCP. Quotas and rate limits follow the current running configuration and are updated on the page and in the interface when they change.
Does the course require a specific deck?
The only real requirement is a tarot deck. The online version is illustrated with the Rider-Waite deck, which the author considers a good starting point because every card has a picture; if you already have a different 78-card deck you may use it, since the principles of tarot work are much the same whatever deck you use.
Terms in this guide
- Self-paced course
- A course structure the learner advances through at their own speed; Learning the Tarot consists of 19 lessons, each with exercises.
- Random draw
- A process in which a program takes cards with equal probability from a full deck; Wenbu's tarot and online hexagram casting work this way, and results should be recorded rather than treated as reproducible.
- Recalculable
- The property of producing the same result again from the same inputs and calculation rules; it applies to deterministic charting, not to random draws.
- Local note
- A record kept in the browser; Wenbu states that visitor records are temporarily stored in the current browser and may be lost if site data is cleared, so export is available as a backup.
See the claim and source notes
Learning the Tarot is a free, self-paced online course of 19 lessons whose only real requirement is a tarot deck.
self-paced series of 19 lessons
Wenbu currently lists tarot draws, local notes and JSON export among features that need no payment or registration.
无需付款或注册
Wenbu states that tarot and online hexagram results are drawn randomly and that reproducible calculation does not mean the interpretation is scientifically validated.
不代表解释已经得到科学验证
Sources and review scope
This AI-assisted Wenbu article passed automated structure checks and a separate editorial model pass. It is not independent expert review or a paid-account product test.
- Wenbu · current free access ↗First-party access and quota documentation · 2026-10-17 · Fetched text
- Wenbu · calculation conventions ↗First-party implementation; not evidence of predictive accuracy · 2026-10-17 · Fetched text
- Wenbu · agent integrations ↗Current first-party product documentation · 2026-10-17 · Fetched text
- Learning the Tarot · Joan Bunning official course ↗Course structure and requirements; do not reproduce copyrighted lessons · 2026-10-17 · Fetched text
- Wenbu · daily card ↗Current first-party product UI; not a personal forecast · 2026-10-17 · Fetched text
Write down one practice question
Write one sentence about what you want to practice, then decide whether to study card meanings with the course or draw and record with Wenbu.
Write down one practice question ↗