Install
$ agentstack add skill-tzengyuxio-skills-wenyan-xianqin ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Wenyan-Xianqin — 改寫為先秦諸子體
將現代中文改寫為先秦諸子散文風格,追求「辭約義豐、一字千金」的極致簡鍊。
代表文本:《論語》《孟子》《莊子》《韓非子》《左傳》《老子》《荀子》
Arguments
$0— (optional) Markdown or plain text file path. If omitted, ask for inline text.
Step 1: Load the Text
- If
$0is provided, read the file. - If no argument, look for the most recently written/edited Markdown file in conversation, or ask user.
Step 2: Scan & Assess
Read the full text. Identify:
- The core insight or teaching (微言大義).
- Modern vocabulary and syntax that need radical compression.
- Whether the text suits aphorism (語錄), disputation (論辯), parable (寓言), or chronicle (記事) form.
Step 3: Apply Pre-Qin Philosophical Patterns
Apply the checklist below. Each item marked [Hard] must be applied; [Soft] items apply when natural.
Checklist: Grammar Patterns (G: 語法句式)
G1. Judgement with 也, Not 是 (也字判斷) [Hard]
Pre-Qin Chinese had NO copula 是. All judgements use the pattern 「X者Y也」「X,Y也」「非X也」. Eliminate every copular 是.
> Before: 仁愛是人類最重要的美德。 > > After: 仁者,人之大德也。
> Before: 這就是治國的根本道理。 > > After: 治國之本也。
G2. Radical Subject Dropping (極致省略主語) [Hard]
Pre-Qin prose drops subjects far more aggressively than any later period. If the subject is inferable from context, DELETE it. A sequence of clauses sharing one subject states it only once — or not at all.
> Before: 孔子說:「我每天都會反省自己三次。」 > > After: 子曰:「日三省。」
> Before: 君子知道什麼是對的,小人只知道什麼有利。 > > After: 君子喻於義,小人喻於利。
G3. Topic-Comment with Zero Copula (主題—述語) [Hard]
Pre-Qin syntax is purely topic-comment. State the topic, then comment directly. No linking verb, no padding.
> Before: 關於政治的要點,最重要的在於贏得人民的信任。 > > After: 政者,民信之。
> Before: 音樂的功能是調和人的情感。 > > After: 樂,和情也。
G4. Classical Conditional & Causal (假設因果) [Hard]
Replace 如果…就 / 因為…所以 with 苟…則、故、是以、以…故. Pre-Qin often omits the conditional marker entirely — juxtaposition implies causation.
> Before: 如果名分不正確,說的話就不會順暢。 > > After: 名不正,則言不順。
> Before: 因為他自己品行端正,所以不需要發布命令,事情自然就能辦好。 > > After: 其身正,不令而行。
G5. Parallel Compression (對偶壓縮) [Hard]
Balance clauses to identical character counts. Pre-Qin parallelism is tighter than any later period — 4:4, 3:3, even 2:2. Each half must mirror the other in structure.
> Before: 知道的人不如喜歡的人,喜歡的人又不如享受其中的人。 > > After: 知之者不如好之者,好之者不如樂之者。
> Before: 在上位的人如果喜歡禮,老百姓就沒有人敢不恭敬。 > > After: 上好禮,民莫敢不敬。
G6. Rhetorical Question Patterns (反問定式) [Soft]
Use 不亦…乎、何…之有、豈…哉、孰…與…、安…、惡…. These are the signature rhetorical forms of Pre-Qin — use them to replace modern rhetorical questions.
> Before: 有朋友從遠方來,難道不是一件很快樂的事嗎? > > After: 有朋自遠方來,不亦樂乎?
> Before: 怎麼可能會有什麼困難的呢? > > After: 何難之有?
Checklist: Vocabulary Patterns (V: 詞彙用字)
V1. Single-Character Words Throughout (單音節詞) [Hard]
Pre-Qin vocabulary is overwhelmingly monosyllabic. Every bisyllabic word must be challenged — most can be reduced to one character:
- 國家 → 國
- 學習 → 學
- 知道 → 知
- 思考 → 思
- 聽到 → 聞
- 看見 → 見
- 說話 → 言
- 擔心 → 憂
- 明白 → 知 / 明
- 喜歡 → 好 / 悅
- 選擇 → 擇
- 努力 → 勉
> Before: 我們應該認真學習並且努力思考。 > > After: 學而思。
> Before: 孔子覺得非常高興。 > > After: 子悅。
V2. Classical Particles (語氣詞) [Hard]
Pre-Qin particles are polysemous, each covering multiple modern functions:
- 也 — judgement, mild assertion, pause
- 矣 — completed action, new situation, exclamation
- 乎 — question, exclamation, preposition
- 哉 — strong exclamation, rhetorical question
- 焉 — "there / in it / how", combines 於+之
- 耳 — "merely, that's all"
- 夫 — sentence-initial topic marker, or final exclamation
> Before: 這件事已經結束了啊! > > After: 事畢矣夫!
> Before: 不過如此罷了。 > > After: 如是而已耳。
V3. Classical Pronouns (代詞) [Hard]
- 我 → 吾 / 余 / 予
- 你 → 汝 / 爾 / 子 / 女
- 他/它 → 其 / 彼
- 自己 → 己
- 誰 → 孰
- 什麼 → 何
- 哪裡 → 安 / 惡 / 焉
- 這 → 此 / 是(指示詞,非繫詞)
- 那 → 彼
Drop pronouns whenever possible (see G2). When retained, use single characters only.
> Before: 你自己不想要的東西,不要強加給別人。 > > After: 己所不欲,勿施於人。
V4. Classical Function Words (虛詞) [Hard]
Core substitutions:
- 的 → 之
- 在 → 於
- 和/跟 → 與
- 用/拿 → 以
- 從 → 自
- 就 → 則 / 乃
- 所以 → 故 / 是以
- 卻/但是 → 而
> Before: 用道德來治理國家,就好像北極星在它的位置上,所有的星星都圍繞著它。 > > After: 為政以德,譬如北辰,居其所而眾星共之。
V5. Classical Negation (否定詞) [Hard]
Pre-Qin negation is richer than any later period:
- 不 — general negation
- 弗 — "not" (with implicit object)
- 未 — "not yet"
- 毋/勿 — prohibitive "do not"
- 莫 — "none / nobody / do not"
- 非 — "is not / not"
- 無 — "there is no"
- 微 — "were it not for" (counterfactual)
> Before: 不要擔心別人不了解你自己。 > > After: 勿患人之不己知。
> Before: 沒有人能夠阻止它。 > > After: 莫之能禦。
V6. Remove ALL Modern Connective Padding (去連詞) [Hard]
Pre-Qin prose uses almost no multi-syllable connectives. 因為…所以, 雖然…但是, 不但…而且, 如果…就 — ALL must go. Replace with juxtaposition, 而, 則, or nothing.
> Before: 雖然學習了但是沒有思考,就會迷惘;雖然思考了但是沒有學習,就會空虛危險。 > > After: 學而不思則罔,思而不學則殆。
V7. Eliminate 把/被 Constructions (去把被句) [Hard]
把 and 被 as grammatical markers did not exist in Pre-Qin. Restructure:
- 被 → passive implied by context, or 見…、為…所…
- 把 → direct verb-object, or 以 + object
> Before: 他被小人陷害了。 > > After: 見陷於小人。
> Before: 我把這件事告訴了他。 > > After: 以告之。
Checklist: Style Patterns (S: 風格修辭)
S1. Aphoristic Brevity (格言凝鍊) [Hard]
Every sentence should feel like it could be carved in stone. If a thought takes more than 8 characters, ask whether it can be split into two 4-character statements. The ideal unit is 4-8 characters.
> Before: 聰明的人喜歡水,仁慈的人喜歡山。聰明的人活潑好動,仁慈的人沉靜安穩。 > > After: 知者樂水,仁者樂山。知者動,仁者靜。
> Before: 一個人做事如果只求差不多,最終會害了自己。 > > After: 苟且者,自賊也。
S2. Parable & Analogy (寓言取譬) [Soft]
Where abstract argument dominates, consider replacing with a concrete image or micro-parable — the signature move of 莊子 and 韓非子.
> Before: 人的知識是有限的,但是世界上需要學習的東西是無限的。用有限的生命去追求無限的知識,是很危險的。 > > After: 吾生也有涯,而知也無涯。以有涯隨無涯,殆已。
> Before: 不要用固定不變的方法去處理不斷變化的情況。 > > After: 刻舟求劍,舟行而劍不行。
S3. Dialogue Markers (對話標記) [Soft]
Pre-Qin philosophical texts are built on dialogue. Use「X曰:」for speech, and the concise call-and-response rhythm of 論語:「問…」「答…」.
> Before: 有學生問孔子什麼是仁。孔子回答說:「愛人。」又問什麼是知。孔子回答說:「知人。」 > > After: 樊遲問仁。子曰:「愛人。」問知。曰:「知人。」
S4. Layered Meaning (微言大義) [Soft]
The best Pre-Qin prose carries meaning beyond the surface. Let compression create ambiguity that rewards re-reading. Do not over-explain.
> Before: 我每天用三件事來反省自己:幫別人做事有沒有盡心?和朋友來往有沒有守信用?老師傳授的學業有沒有認真複習? > > After: 日三省吾身:為人謀而不忠乎?與朋友交而不信乎?傳不習乎?
Checklist: Register Markers (R: 時代標記)
R1. NO Multi-Syllable Conjunctions (禁多音節連詞) [Hard]
This is the single most important register marker. Pre-Qin had NONE of these: 因為、所以、雖然、但是、而且、或者、如果、即使、不但、因此、然而、可是. Their presence instantly breaks the register. Every one must be eliminated.
> Before: 因為教育不普及,所以很多人不知道禮義。然而如果能夠推行教化,即使是普通百姓也可以成為君子。 > > After: 教不行,民不知禮義。苟行教化,則庶人可為君子。
R2. Classical Time & Place (時地詞) [Soft]
- 那時候 → 是時 / 當是時
- 後來 → 既而
- 從前 → 昔 / 昔者
- 現在 → 今
- 以前 → 先 / 曩
- 將來 → 後
> Before: 從前有一個農夫在田裡耕作。 > > After: 昔者農人耕於田。
R3. Sentence Rhythm: 4-Character Base (四言為基) [Hard]
Pre-Qin prose gravitates toward 4-character phrases as the default rhythm unit, occasionally expanding to 5 or 6, contracting to 2 or 3. This 4-char pulse is the heartbeat of the style.
Hard rule: each clause defaults to 4 characters. Clauses over 6 characters need strong justification. Clauses over 8 characters should almost never appear.
> Before: 如果自己的行為端正,不用下命令事情也會被執行;如果自己的行為不端正,即使下了命令也沒有人會服從。 > > After: 其身正,不令而行;其身不正,雖令不從。
> Before: 差不多先生的相貌,跟你跟我都差不多。(14字子句) > > After: 其貌與眾同。(5字)
Step 4: Rewrite
Apply all checklist items to produce a complete rewrite. Follow these principles:
Rewriting Principles
- 一字千金 — Every character must earn its place. If it can be removed without losing meaning, remove it.
- 辭約義豐 — Maximum meaning in minimum characters. Target: output should be 30-50% of the input's character count. If the output exceeds 50%, compression is insufficient — go back and cut harder.
- 述而不論 — State; do not explain. Trust the reader to draw the inference.
- 正言若反 — The most powerful statements often take paradoxical form (老子's signature).
- 取譬明理 — When logic alone is dry, one concrete image outweighs ten abstract arguments.
- 寓莊於諧 — Serious ideas can ride on playful or absurd examples (莊子's method).
Target Quality: Complete Example
Below is a short passage rewritten in the target register — this is the level of compression and rhythm to aim for:
> Modern (68 chars): 他是一個做事只求差不多的人。別人都說他看得開、想得通,認為他很有修養。但其實他只是懶惰和馬虎,最後因為這個態度而喪命。 > > 先秦 (24 chars): 苟且者也。人以為達,實怠也。卒以此死。
Note: 24/68 ≈ 35% — within the target compression ratio.
Step 5: Authenticity Audit
Pass 1 — Pattern Re-scan
Re-read the rewritten text against the checklist above. Check for:
- Any surviving copula 是 (most common failure)
- Any multi-syllable conjunctions (因為, 所以, 雖然, 但是, etc.)
- Any 把/被 constructions
- Any bisyllabic words that could be monosyllabic — go sentence by sentence: if a sentence of 8 chars contains 2+ bisyllabic words, it has not been compressed enough
- Clauses exceeding 8 characters that could be split
- Character count check: count the total characters of the output vs input. If the ratio exceeds 50%, the rewrite needs further compression
Pass 2 — Read-Aloud Test
Read the whole piece aloud (mentally). Ask:
- Does it sound like something from 論語 or 莊子, or does it sound like a modern student imitating classical Chinese?
- Is the rhythm rooted in 4-character phrases with natural variation?
- Are subjects dropped aggressively enough? (If you can point to who is doing what in every clause, you have not dropped enough.)
- Does it feel carved, not written?
Fix any remaining issues.
Step 6: Output
- Save to new file with
_xianqinsuffix (e.g.,article.md→article_xianqin.md). If user explicitly requests in-place edit, overwrite. - Show 3-5 most representative before/after pairs.
- State which patterns were most heavily applied.
Important Notes
- This is the most extreme compression of all classical Chinese styles. Meaning WILL be compressed and some nuance may be lost. This is inherent to the style, not a defect.
- For most practical purposes (essays, proposals, formal writing),
wenyan-guwen(Tang-Song prose) is more appropriate. Use this skill only when the Pre-Qin philosophical register is specifically desired. - This is a style transformation tool, not a content generator. Preserve the original meaning to the extent the extreme compression allows.
- Do NOT add teachings, parables, or philosophical content not present in the original.
- When the source contains modern technical terms with no classical equivalent (e.g., 半導體, 演算法), keep the term and transform the surrounding prose.
- Pre-Qin prose ≠ poetry. Do NOT force rhyme or uniform line length. The rhythm is prose rhythm — regular but not rigid.
- 先秦諸子 each have distinct voices: 論語 is terse dialogue, 莊子 is imaginative parable, 韓非子 is sharp argument, 左傳 is laconic narrative. Match the sub-style to the source text's nature when possible.
- Pair with
wenyan-guwen(Tang-Song essay),wenyan-hanwei(historical narrative), orwenyan-huaben(vernacular fiction) for other classical styles.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: tzengyuxio
- Source: tzengyuxio/skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.