去AI味改写·论文降重|AI文本自然化
Academic writing style toolkit, bilingual CN/EN: paper rewriting with AI-flavor self-check and style naturalization (scan reports are detection only), deterministic cleanup (model artifacts, punctuation, filler phrases), integrity guardrails (numbers, DOIs, PMIDs, terminology must survive untouched), and agent-guided revision for clearer, more natural academic prose. Built for scholars whose formal writing reads stiff, templated or machine-flavored — especially non-native (ESL) authors. Skip when: pure grammar polish, full-text translation, or rewriting unrelated to academic writing — use a dedicated tool. Includes AIGC-disclosure compliance checks (China 2025-09 labeling rules) and academic-integrity guardrails: this tool improves writing quality for self-review; it does not help misrepresent authorship, conceal required AI disclosure, or defeat integrity review. 100% local, zero upload, Python stdlib only. Related: paper-polisher-pro (broad polishing), pubmed-verifier, cite-holmes, academic-figures.
skillhub install @indiv-sorsor/humanize-ai-text
README 文档
16,125 字Academic Style Toolkit (humanize-ai-text)
Bilingual (CN/EN) style naturalization for academic & medical writing: find stiff, templated or machine-flavored patterns, clean mechanical debris, revise for clarity and natural register — with integrity guardrails on every step. Reading map (30-second routing; equivalent of references/INDEX.md) — take what you need:
- Start / troubleshoot: quick start · pitfalls + errors
- Flow / scenarios: deeprewriteguide · best_practices · CI/agent → automation
- Limits / compliance: honest_boundaries · permissions · compliance
- Reference / tools: api · pdf_confidence · calibration · customization
Two layers in one line: mechanical cleanup (transform/pipeline, deterministic) vs quality revision (agent per style guide) — run pipeline first, read the report, then decide what to deep-revise.
Routing in one line: grammar/fluency/journal-format polish → use a polisher (paper-polisher-pro); de-templating, model-residue cleanup, AI-flavor style self-check and rewriting → this tool. Need both? Polish first, then run this pipeline.
Feature status (v5.16.0)
⚠️ rows = experimental (may reject valid files; review output). The full per-generation capability list has moved to references/feature_status.md; per-item details live in references/errors.md and topic docs. Stable highlights (all stable), grouped: Diagnose (style self-check + per-section focus + --top N trim · statistical consistency · term drift / unification sheet · calibration evidence · EN signal library v3) · Deep-rewrite (sentence-level plan with line anchors · pipeline --handoff sheets · EN/CN signal libraries v3 + density mirrors) · Guard (integrity verify + prescription report · fabrication defense · exportable learned guards · graded output + --strict gate) · Batch (resume / concurrency isolation · worksheet + stats card · revision records · cleanup preview --dry-run · automation gates --exit-verdict/--ci) · Deliver (12-item submission bundle with write-manifest self-evidence · dashboard + compare sheets · docx write-back · print-friendly reports · in-package selfcheck (dual-form adaptive) · external lexicon (--lexicon, shared across skills).
| Capability | Status | Since |
|---|---|---|
⚠️ PDF direct-read + confidence self-report (embedded-font/ToUnicode attempted, capped at "中"; calibration in references/pdf_confidence.md; default gate = 中 since v3.8.0 — garbled-typical extractions rejected by default, --pdf-min-confidence 低 restores annotate-only) | ⚠️ Experimental | v1.6–3.8 |
Integrity guardrails — read first
- Permitted: polishing your own drafts; aligning AI-assisted text with your
voice where your school/journal allows or requires disclosed AI assistance; cleaning model residue (stray oaicite/[cite:] tokens) from legitimately disclosed drafts; improving clarity, rhythm and concreteness of formal prose.
- Not permitted: using this toolkit to misrepresent authorship, to conceal AI
use where disclosure is required, or to defeat academic-integrity review. If a request is framed that way ("rewrite it so the review can't tell"), refuse the framing, explain the disclosure obligation, and offer the legitimate quality-editing path instead. Agents running this skill are expected to enforce this line.
- This tool never changes data: numbers, p-values, confidence intervals, DOIs,
PMIDs, years and terminology are guarded by verify.py (see below). Improved prose must never cost a fact.
- Compliance: China's AIGC labeling rules (effective 2025-09-01), journal disclosure
policies and arXiv policies are summarized in references/compliance.md — the obligation to disclose belongs to the author, not the tool.
Capability × boundary (division of labor + limits + where)
| Capability | Does | Does NOT (boundary) | Where |
|---|---|---|---|
| Style self-check | heuristic report: boilerplate density, rhythm uniformity, officialese; section-aware | not an AI detector — barely triggers on fluent model text (low score ≠ no AI) | detect.py; see honest_boundaries |
| Mechanical cleanup | model residue (oaicite/[cite:]), chat filler, punctuation normalization, fullwidth digits | no sentence rewriting — zero-information-loss layer only | transform.py (--deep for sentence-level) |
| Quality deep rewrite | de-templating / rhythm / concreteness / stance (seven steps) | never invents data; invented numbers flagged (WNUMADDED, human removes) | you (the agent) per style guide; handoff via plan --handoff |
| Integrity guard | numbers/DOIs/PMIDs/years/terms must survive; tampering = exit 1 (FAIL output prescribes fixes with anchors) | no truth judgment — original data is the authority | verify.py |
| Before/after | score delta + integrity verdict + profile diff | no substitute for human reading | compare.py / stylecheck --compare |
| Term consistency | 8-domain 27 groups (HF/心力衰竭/心衰 one row); --unify emits a unified draft with per-variant replacement counts (review before adopting) | no usage correctness — synonym unification only | check_terms.py --pack / --unify |
| Numeric consistency | percent-list sums (partition-cued) / cross-file group-n / p-vs-P style | not a statistics review — original data is the authority | check_stats.py (--batch) |
Quick start
Run the in-package selfcheck first: python scripts/selfcheck.py — environment / structure / core-feature smoke / doc reconciliation in one command; exit 0 means installed correctly (ships with the published package; --json for CI/agents).
How to ask for this in plain words (EN digest of the full phrase table in SKILLZH.md): "remove the AI flavor from this paper" / "this reads too templated" / "clean up oaicite and [cite:] residue" / "tell me which sentences to fix first" / "check my rewrite broke no numbers" / "give me a style profile" / "stop flagging this legitimate term" — each maps to the pipeline, plan.py, transform.py, verify.py+compare.py, stylecheck.py and learnguards.py respectively. No agent is required — every capability below runs directly in a terminal. Every script has built-in --help (full flag reference) and --version; quick flag semantics:
| Flag | Meaning |
|---|---|
--profile general | non-academic text: down-weight formulaic/boilerplate signals |
--deep | deterministic sentence-level transforms (canned openers, not-only merge) |
--exit-verdict | machine exit codes for CI/agent gating (see below) |
--structure | section-aware scoring (IMRaD) |
--step-timeout SEC | per-step wall-clock budget (default 120) |
--terms FILE | term list for the integrity guard (build via extract_terms.py) |
--unify SHEET | check_terms: emit a term-unified draft (advisory — review the replacement list) |
--section abstract | detect: diagnose one section in isolation (the abstract is what reviewers read first) |
-o revised.docx | -o on pipeline/transform/compare writes Word when the path ends in .docx (plain paragraphs; merge manually for complex layouts) |
One command (self-check → cleanup → revision brief → integrity guard):
python scripts/pipeline.py draft.txt -o out.txt --terms terms.txt
- Visualization / batch:
--html report.htmlon detect/pipeline/compare for a
self-contained report (compare adds sentence-level diff); --batch dir scans a whole directory (pipeline writes a CSV; the batch -o summary stays CSV — no docx write-back for the table); -o on pipeline/transform/compare ending in .docx is written back as Word (plain paragraphs — closes the loop with .docx direct-read; tables/images/styles are not rebuilt, merge complex layouts manually).
- Section-aware:
detect --structurescores abstract/intro/methods/results
separately; methods/results phrasing is genre convention — normalized in the aggregate only, raw per-section scores always shown.
- Rewrite plan:
plan.py draft -o plan.mdranks sentences (P0 queue +
budget projection); add --handoff for a per-sentence operation sheet.
- Inputs:
.txt/.mddirect (UTF-8/UTF-16〔with BOM〕/GBK auto-transcoded, v4.0),.docxdirect, PDF experimental (default safe gate = medium since v3.8;低restores annotate-only; references/pdf_confidence.md). - Term list:
python scripts/extract_terms.py draft.txt -o terms.txt(confirm
by hand); any run accepts --suggestions path.md for a revision worksheet.
- Non-academic text: add
--profile generalto down-weight formal boilerplate.
Or run the steps individually (from the skill's own directory, or use absolute paths):
python scripts/detect.py draft.txt # style-pattern self-check (-j JSON, -s score only)
python scripts/transform.py draft.txt -o step1.txt
# ... agent quality revision of step1.txt per the style guide ...
python scripts/verify.py draft.txt step2.txt --terms terms.txt
python scripts/compare.py draft.txt step2.txt
Scenario-organized best practices (journal submission, thesis, revision letters, batch, CI gates, false-positive handling, CN punctuation): references/best_practices.md. Worked end-to-end examples: references/examples.md. Unified exit codes, violation categories and remedies: references/errors.md. Centralized common-mistakes list (27 items): references/pitfalls.md — read it before your first real run. Python API reference: references/api.md. FAQ: references/faq.md.
- Revision tracking:
--track baserecords every mechanical-layer change to
base.md + base.json (auditable: rule, count, deleted sentences, flagged).
- PDF (experimental): text PDFs incl. embedded-font/ToUnicode since v3.0.0 —
extraction attempted, confidence capped at "中" with a spot-check note; CJK CID files still rejected by the garbage probes with export guidance. Scale: 800K-char chunks (2K overlap) · 50MB cap · whole-document integrity guard.
Agent invocation protocol
When invoked, decide the path first, then run it:
- Trigger words: 写作风格自查 / 论文改写润色 / 去模板腔 / 翻译腔清理 /
学术改写 / 段落改写 / 表达优化 / style self-check / style naturalization / naturalize academic writing / de-templating / academic rewriting / passage rewrite / expression polish / translationese cleanup (= Europeanized syntax: stacked passives, nominalization chains, long attributives — see styleguideen steps 1/3) / make my writing sound less machine-written / reduce the templated tone / clean up model residue or stray citation tokens / fix stiff formal prose / polish my draft for submission without changing the data / check that numbers and citation IDs survived editing → run the pipeline above.(中文变体触发词见 SKILLZH.md——CH 分寸词不入英文文件; 数字自洽/术语混用类请求路由到 checkstats.py / check_terms.py,见下) (Parameter cheat sheet lives in Quick start; exit-code tables in references/errors.md — both are linked from here to avoid hunting across sections.)
- Disambiguation (this tool vs a polisher): 「论文改写润色」 here means
style naturalization / de-templating (removing the machine flavor from academic prose). Decision rule — route to a polisher (paper-polisher-pro) when the ask is only about language correctness/fluency: 「帮我改下语法」 「按期刊风格改写摘要」; route to this toolkit when it is about templated tone or model residue: 「这篇读起来像 AI 写的」「文里有 [cite: 1] 这种残留」; when both apply, polish first, then run this pipeline — neither substitutes for the other.
- Term inconsistency (abbreviation vs full form, mixed synonyms) →
check_terms.py file --pack <domain>; to emit a unified draft directly add --unify output.txt (each group unified to its dominant variant, per-variant replacement counts surfaced — advisory output, review before adopting, then run verify; v4.4.0); --list-packs shows all eight domain packs.
- Numbers that don't agree (percent list not summing to 100, same group
with two different n, p/P style mix) → checkstats.py file ( exit codes match checkterms: 0 consistent / 1 issues found). Chaptered manuscripts: check_stats.py --batch 目录 aggregates group-n across files (abstract-vs-methods mismatches, v3.3.0). Or run everything at once: pipeline.py draft --check-stats.
- User wants to know what to fix first →
plan.py draft.txt -o plan.md
(P0 sentence queue + budget projection); add --handoff to emit the per-sentence instruction sheet (guards + acceptance commands, v3.0.0).
- User wants quantified acceptance of a rewrite → `stylecheck.py 原稿 改稿
--compare` (style profile delta; triggers: 风格画像 / 量化验收 / 深改验收 / style profile / acceptance check).
- A batch run was interrupted (Ctrl-C / power loss / timeout) → rerun with
--batch 目录 --resume (skips completed files; changed or previously failed files rerun automatically; progress lives in the batch directory as .prbatchstate.json).
- Scenario shortcuts (v3.8.0 consolidation): CI/batch gating →
--exit-verdict (--ci alias; 0 clean / 3 deep-revise / 4 residue / 1 integrity fail, pipeline-only); false-positive remediation → learn_guards.py from-text 术语 样本.txt; where to start deep revision → pipeline 原稿 -o 改稿 --handoff 交接块.md (one command, no separate plan run).
- Multi-chapter term unification →
check_terms.py --batch 目录(cross-file
variant drift: each file internally consistent, the set inconsistent).
- A scan flags a legitimate term (false positive) → don't edit the pattern
files; persist a guard instead: learn_guards.py from-text 术语 样本.txt (or add) — the guard survives upgrades and applies to every later scan.
- User only wants a verdict on an existing rewrite → `pipeline.py draft.txt
--rewrite rewritten.txt --terms terms.txt` (skip cleanup).
- **User asks to conceal AI use, misrepresent authorship, or defeat integrity
review** → refuse the framing, point to references/compliance.md, offer the legitimate quality-editing path instead. Do not run the toolkit toward that end.
- Exit codes:
verify.pyis the strict contract holder —0integrity PASS,
1 integrity FAIL (fix the rewrite, never the guard), 2 usage/file error. detect/transform/compare/pipeline report the verdict in their output and always exit 0 on completed runs, 2 on usage/file errors. For agent/CI branching, add --exit-verdict: machine exit codes 0/3/4 on detect/compare, full 0/1/3/4 on pipeline — see Automation & CI integration below. Every script supports --version.
Automation & CI integration
- Default (human flow):
verify.pyowns exit codes 0/1/2; other scripts write
verdicts into their reports — do not branch on their default exit codes.
- Machine mode: detect / compare / pipeline take
--exit-verdict(alias--ci),
five codes 0/3/4/1/2; pipeline.py is the only full gate (1/3/4). The one-liner and the JSON field tables: automation · errors section 1.
The workflow
- Self-check:
detect.py draft.txt— a heuristic report of style patterns
(canned phrases, uniform sentence rhythm, boilerplate transitions, model artifacts). The report is for the author's own review; the score is a local heuristic, not an official measurement of anything.
- Mechanical cleanup:
transform.py draft.txt -o step1.txt— strips stray
model artifacts (oaicite, turn0search, [cite: 1], grokcard, attachedfile), leftover chatbot pleasantries, markdown residue, and a safe list of filler phrases; CN halfwidth punctuation is normalized to fullwidth (decimals protected). -a adds em-dash reduction and empty-opener removal. --deep adds sentence-level deterministic transforms — canned opener deletion, 「不仅X,而且Y」→「X,且Y」 merging — zero-information-loss only, every operation logged to --track — the log is the audit trail. Text with no hits passes through with zero edits (fullwidth alnum normalization is the only exception; per-replacement counts are re-runnable via --track, so any claim here can be verified locally).
- Quality revision (the real work): optionally rank the work first —
python scripts/plan.py draft.txt -o plan.md gives a sentence-level P0 queue (worst offenders with section, category and advice) so deep effort lands where the score lives; add --handoff to emit a per-sentence operation sheet (original sentence + directive + guard rules + acceptance commands) that a rewriting agent can follow verbatim. Then read the guide for the text's language —
- CN:
references/styleguidezh.md— structural de-templating → jargon
cleanup → rhythm → concreteness → stance → integrity red lines
- EN:
references/styleguideen.md— smaller words → fewer significance
frames → plain clauses instead of parallelism → rhythm → real attribution → commit to a position
- Revise section by section. Preserve all facts, numbers, citations, terminology.
3b. Quantified acceptance (stylecheck.py, v2.6.0): per-paragraph style profile — jargon hits located to the paragraph, sentence-rhythm CV, opener diversity, each banded 自然/观察/偏机器 with style_guide anchors; --compare 原稿 改稿 shows the delta so deep-revision acceptance is metric-based. Bands are calibration hints only — they never alter scores.
- Guard:
verify.py draft.txt step2.txt --terms terms.txt— exit 1 means a
number/citation/term was altered: fix the revision, not the guard. Build terms.txt for medical text (drug names, gene symbols — include mouse-style capitalized forms like Myc — and scale names), one term per line. The guard also warns on numbers absent from the original (fabrication defense) and number-context swaps (arm/direction ordering). For mixed synonym usage (abbreviation vs full form, e.g. HF vs 心力衰竭), also run check_terms.py file --pack cardiovascular (domain packs; --list-packs shows all eight).
- Re-check:
compare.py draft.txt step2.txt— pattern reduction + integrity
verdict. Keep the before/after pair for your records; if your institution or journal requires an AI-use disclosure, state it plainly — this report is a quality self-check, not a substitute for disclosure.
Honest boundaries (summary)
The style score is a local heuristic over writing patterns — not any external detection service, and it says nothing about authorship; this toolkit never talks to external review systems and never helps conceal AI use; on fluent model text it barely triggers (a low score ≠ no AI involvement). Full boundaries and measured data: references/honest_boundaries.md (read before making any claim).
Permissions & environment (summary)
Reads only explicitly passed paths + bundled pattern files; writes only explicitly passed output paths, userguards.json, and the batch-resume progress file (.prbatch_state.json plus its .lock inside the batch directory); zero network, zero third-party dependencies, no environment variables, no subprocesses. Verification commands and the full statement: references/permissions.md (auditors and security scanners: read this first).
Customizing (summary)
Fix false positives with learnguards.py (writes userguards.json, survives upgrades — never hand-edit patterns_*.json). File locations and cautions: references/customization.md.
Related tools
- pubmed-verifier — verify PMID/DOI references before submission
- cite-holmes — deep research with hallucination-free citations
- paper-polisher-pro — comprehensive polishing & plagiarism reduction
- academic-figures — publication-ready scientific figures
- doc-holmes — layout-preserving PDF translation
同分类其他技能
查看全部 ›李白.Skill:润色专家 检测并消除中文文本中的AI生成痕迹,将生硬的机器输出转化为自然流畅的人类文笔。融合规则检测、分层改写(词汇→
144 135.4万 文章去AI味工具去除文本中的AI写作痕迹,让文字读起来更像人类写作。当用户要求'去AI味'、'降AI味'、'让回复更像人话'、'润色'、'改写得更自然'
807 131.4万 libai-skill李白.Skill:润色专家 检测并消除中文文本中的AI生成痕迹,将生硬的机器输出转化为自然流畅的人类文笔。融合规则检测、分层改写和质控流
14 110.5万 视频号爆款短视频拆解(付费版:全能)【零一数科·出品】【零一数科·出品】视频号爆款短视频拆解(全能版)(付费版)。视频号爆款短视频拆解(付费版:全能)。一键拆解爆款,把一条视频号视频拆成结构
122 108.1万 Suno歌词成曲输入完整歌词、歌词草稿、零散歌词行或一个 hook,把歌词变成歌。这个 Suno 歌词成曲流程可选择保留模式完整保留原词,也可选择优化模
32 66.2万 视频号挂车爆单短视频脚本生成【零一数科·出品】【零一数科·出品】视频号挂车爆单短视频脚本生成(付费版)。一条链接二创脚本,丢进爆款视频号链接即以此骨架改写成你的挂车脚本,或按方向/目
40 63.8万