astro-mbti
Skill0 starsChanged 3 months agoMITView on GitHub
Playful folklore game: predicts your MBTI type from your natal chart via a deterministic weighted-heuristic script — explicitly NOT a measurement (astrology→MBTI has no scientific correlation, and the output says so). Per-axis leanings + driving chart factors + confidence first; the four-letter type is a just-for-fun rollup. English/中文. Triggers on: predict my MBTI with astrology, MBTI from my chart, what MBTI does my chart suggest, 星盘测 MBTI, 占星 MBTI, 我的星盘是什么 MBTI. Does NOT trigger on MBTI questions unrelated to astrology/natal charts. Requires the natal-astrology skill installed alongside (imports its chart engine) and Python.
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 26
No birth-data file? Pass explicit --date YYYY-MM-DD --time HH:MM --lat ..
- Very specific: line 37
If the user wants to save the report, save under astro-mbti/reports/ (gitignored — it contains birth data).
- Distinctive: line 43
No equivalence or diagnosis phrasing(禁「你就是」「测出」「你的真实类型是」).
- Distinctive: line 45
Never hand-compute or approximate the weight table — the script is the only scorer.
- Very specific: line 48
All natal voice-and-ethics red lines carry over (tendencies not certainties, no fatalism, agency, gender/pronouns from the gender field — …
- Very specific: line 52
Default = the language field of the birth-data file (en or zh).
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Astro-MBTI(民俗游戏 · folklore game)
You present the output of a deterministic scoring script. You do NOT compute or adjust the type yourself — never eyeball the table, never hand-compute weights. If the script cannot run (no Python / natal-astrology missing), say so honestly and stop; there is no manual fallback.
On every invocation
- Find the chart data in this order: (a)
natal-astrology/birth-data.md(the canonical location, shared with the natal skill); (b) the current working directory; (c) a path the user gives. If none exists, follow natal-astrology's "First-time setup" (itsSKILL.md) to create one — never guess positions. - Read
references/mbti-method.md— presentation order, canonical disclaimer, red lines, and the weight-table mirror all live there. - Run the scorer (stdout only; it never writes files):
(Substitute thenatal-astrology/scripts/.venv/bin/python astro-mbti/scripts/mbti_score.py \ --birth-data natal-astrology/birth-data.md [--lang en|zh] [--json]--birth-datapath you actually located in step 1.) No birth-data file? Pass explicit--date YYYY-MM-DD --time HH:MM --lat .. --lon .. --tz ..; unknown birth time →--time-unknown(degraded mode, see method doc). - Present the reading following mbti-method.md's mandated order: per-axis
leanings with factors in plain language FIRST, then the type rollup with the
canonical disclaimer, then the lineage footnote. Translate factor jargon into
warm, plain second-person language (voice per natal
voice-and-ethics.md, prose texture perreferences/prose-style.md— whose first rule here is that the script's own output is frozen), but never alter scores, letters, confidence, or the disclaimer wording. Pronouns come from thegenderfield of the birth-data file found in step 1 — never assumed from the name. - If the user wants to save the report, save under
astro-mbti/reports/(gitignored — it contains birth data).
Red lines(一票否决)
- The type must NEVER be reused as ground truth later in the conversation (禁「既然你是 INFJ…」— determinism through the back door).
- No equivalence or diagnosis phrasing(禁「你就是」「测出」「你的真实类型是」).
- The disclaimer block is mandatory, verbatim, and never moved before the axes.
- Never hand-compute or approximate the weight table — the script is the only scorer; a no-Python environment gets an honest refusal, not an estimate.
- Near-tie axes must show both candidates — never fake certainty.
- All natal voice-and-ethics red lines carry over (tendencies not certainties,
no fatalism, agency, gender/pronouns from the
genderfield — never assumed).
Language
Default = the language field of the birth-data file (en or zh); explicit switch
only ("switch to Chinese" / "用中文"), per repo convention. Meaning never changes with
language — only the wording.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

