polish-language

Name: polish-language
Author: Aperivue

34installs53forks

How to install polish-language?

aperivue/medsci-skills/polish-language

$npx -y skills add Aperivue/medsci-skills --skill polish-language

Installs into the current project.

Summary

Academic English consistency linting and non-native (ESL) language polish for medical manuscripts. Deterministically flags abbreviation define-once violations, US/UK spelling drift, hyphen-vs-en-dash numeric ranges, P/p case, hyphenation variants, small-number style, and value/un

Files · 1

View on GitHub

SKILL.md

1	# Polish-Language Skill
2
3	You help a medical researcher tighten a manuscript's **mechanical language
4	consistency and clarity** before circulation or submission — the copy-editor
5	pass that content-focused skills skip. The author is frequently a non-native
6	(ESL) English writer, so clarity edits must preserve the formal academic
7	register while never touching facts.
8
9	## Communication Rules
10
11	- Manuscript content and edits in English.
12	- Converse with the user in their preferred language.
13	- Report issues first; only edit after the user approves (see gates below).
14
15	## Scope boundary (what this skill is, and is not)
16
17	\| Concern \| Skill \|
18	\|---\|---\|
19	\| Mechanical consistency + ESL clarity (this skill) \| polish-language \|
20	\| Removing AI writing tells / de-AI \| `humanize` (it explicitly does not do general copy-editing) \|
21	\| Drafting or restructuring content \| `write-paper` \|
22	\| Reporting-guideline item compliance (STROBE, CLAIM, …) \| `check-reporting` \|
23	\| AI-search-engine optimization (GEO) \| `academic-aio` \|
24	\| Reference formatting / citation integrity \| `manage-refs`, `verify-refs` \|
25
26	This skill never rewrites scientific claims, changes numeric values, edits
27	citations, or judges study quality. It only standardizes house style and
28	improves sentence-level clarity with explicit user approval.
29
30	## Inputs / Outputs
31
32	- Input: a manuscript or section (Markdown / plain text).
33	- Output: (1) a deterministic consistency report, and (2) — only after a
34	user gate — a clarity-polished revision with a change log limited to style.
35
36	## Workflow
37
38	### Phase 1: Deterministic consistency lint (no LLM judgement)
39
40	Run the bundled deterministic linter — it reports, never edits:
41
42	```bash
43	python3 scripts/lint_consistency.py path/to/manuscript.md
44	# add --strict to exit non-zero when any issue is found (CI / pre-submission gate)
45	```
46
47	It flags seven families, each with line numbers and a per-category + total
48	count:
49
50	1. Abbreviations — used-before-defined, defined-but-unused, defined-twice,
51	used-but-never-defined (define-once discipline).
52	2. Spelling — mixed US/UK variants (analyze/analyse, tumor/tumour, …);
53	reports the minority side against the document's dominant variant.
54	3. Numeric ranges — hyphen between numbers where an en-dash belongs
55	(`5-10` → `5–10`).
56	4. p-values — mixed `P`/`p` case; impossible `P = 0.000`.
57	5. Hyphenation / terminology — variant forms of one term
58	(follow-up / followup / "follow up").
59	6. Small numbers — single digits 1–9 written as digits in prose.
60	7. Units — missing space between value and unit (`5mg` → `5 mg`).
61
62	Present the report to the user. The linter output is the source of truth for
63	what is mechanically wrong; do not invent additional "issues" from memory.
64
65	### Phase 2: Triage with the user (gate)
66
67	Walk the user through the report. Some flags are author choices (a journal may
68	mandate UK spelling, or digits for all numbers). User approval is required
69	before any edit — confirm per category which to apply and which to keep. Record
70	the decisions; do not auto-apply.
71
72	### Phase 3: Apply mechanical fixes (style-only)
73
74	For each approved category, apply the deterministic fix with `Edit`:
75	- standardize spelling to the chosen variant,
76	- replace numeric-range hyphens with en-dashes,
77	- normalize `P`/`p` and fix `P = 0.000` to the reported inequality,
78	- unify hyphenation, spell out small numbers, add value/unit spaces,
79	- define each abbreviation once at first use; remove redundant redefinitions.
80
81	Re-run `lint_consistency.py` after editing — the count should drop to the
82	issues the user chose to keep. This re-run is the verification gate.
83
84	### Phase 4: ESL clarity polish (optional, gated, style-only)
85
86	If the user requests a clarity pass, improve readability sentence by sentence
87	while preserving meaning, register, numbers, and citations:
88	- split run-on sentences; fix article (a/an/the) and preposition usage;
89	- correct subject–verb agreement and awkward non-native phrasings;
90	- prefer active voice only where it does not change emphasis or claims.
91
92	Show each proposed change as a before/after diff and get user review before
93	writing. If a sentence's meaning is even slightly uncertain, leave it and ask —
94	do not guess. Never merge, add,

People who liked this also viewed

6 picks

Type

polish-language

34installs53forks

View on GitHub

How to install polish-language?

aperivue/medsci-skills/polish-language

$npx -y skills add Aperivue/medsci-skills --skill polish-language

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Polish-Language Skill
2
3	You help a medical researcher tighten a manuscript's **mechanical language
4	consistency and clarity** before circulation or submission — the copy-editor
5	pass that content-focused skills skip. The author is frequently a non-native
6	(ESL) English writer, so clarity edits must preserve the formal academic
7	register while never touching facts.
8
9	## Communication Rules
10
11	- Manuscript content and edits in English.
12	- Converse with the user in their preferred language.
13	- Report issues first; only edit after the user approves (see gates below).
14
15	## Scope boundary (what this skill is, and is not)
16
17	\| Concern \| Skill \|
18	\|---\|---\|
19	\| Mechanical consistency + ESL clarity (this skill) \| polish-language \|
20	\| Removing AI writing tells / de-AI \| `humanize` (it explicitly does not do general copy-editing) \|
21	\| Drafting or restructuring content \| `write-paper` \|
22	\| Reporting-guideline item compliance (STROBE, CLAIM, …) \| `check-reporting` \|
23	\| AI-search-engine optimization (GEO) \| `academic-aio` \|
24	\| Reference formatting / citation integrity \| `manage-refs`, `verify-refs` \|
25
26	This skill never rewrites scientific claims, changes numeric values, edits
27	citations, or judges study quality. It only standardizes house style and
28	improves sentence-level clarity with explicit user approval.
29
30	## Inputs / Outputs
31
32	- Input: a manuscript or section (Markdown / plain text).
33	- Output: (1) a deterministic consistency report, and (2) — only after a
34	user gate — a clarity-polished revision with a change log limited to style.
35
36	## Workflow
37
38	### Phase 1: Deterministic consistency lint (no LLM judgement)
39
40	Run the bundled deterministic linter — it reports, never edits:
41
42	```bash
43	python3 scripts/lint_consistency.py path/to/manuscript.md
44	# add --strict to exit non-zero when any issue is found (CI / pre-submission gate)
45	```
46
47	It flags seven families, each with line numbers and a per-category + total
48	count:
49
50	1. Abbreviations — used-before-defined, defined-but-unused, defined-twice,
51	used-but-never-defined (define-once discipline).
52	2. Spelling — mixed US/UK variants (analyze/analyse, tumor/tumour, …);
53	reports the minority side against the document's dominant variant.
54	3. Numeric ranges — hyphen between numbers where an en-dash belongs
55	(`5-10` → `5–10`).
56	4. p-values — mixed `P`/`p` case; impossible `P = 0.000`.
57	5. Hyphenation / terminology — variant forms of one term
58	(follow-up / followup / "follow up").
59	6. Small numbers — single digits 1–9 written as digits in prose.
60	7. Units — missing space between value and unit (`5mg` → `5 mg`).
61
62	Present the report to the user. The linter output is the source of truth for
63	what is mechanically wrong; do not invent additional "issues" from memory.
64
65	### Phase 2: Triage with the user (gate)
66
67	Walk the user through the report. Some flags are author choices (a journal may
68	mandate UK spelling, or digits for all numbers). User approval is required
69	before any edit — confirm per category which to apply and which to keep. Record
70	the decisions; do not auto-apply.
71
72	### Phase 3: Apply mechanical fixes (style-only)
73
74	For each approved category, apply the deterministic fix with `Edit`:
75	- standardize spelling to the chosen variant,
76	- replace numeric-range hyphens with en-dashes,
77	- normalize `P`/`p` and fix `P = 0.000` to the reported inequality,
78	- unify hyphenation, spell out small numbers, add value/unit spaces,
79	- define each abbreviation once at first use; remove redundant redefinitions.
80
81	Re-run `lint_consistency.py` after editing — the count should drop to the
82	issues the user chose to keep. This re-run is the verification gate.
83
84	### Phase 4: ESL clarity polish (optional, gated, style-only)
85
86	If the user requests a clarity pass, improve readability sentence by sentence
87	while preserving meaning, register, numbers, and citations:
88	- split run-on sentences; fix article (a/an/the) and preposition usage;
89	- correct subject–verb agreement and awkward non-native phrasings;
90	- prefer active voice only where it does not change emphasis or claims.
91
92	Show each proposed change as a before/after diff and get user review before
93	writing. If a sentence's meaning is even slightly uncertain, leave it and ask —
94	do not guess. Never merge, add,

People who liked this also viewed

6 picks

Type