analyze-stats

Name: analyze-stats
Author: Aperivue

47installs53forks

How to install analyze-stats?

aperivue/medsci-skills/analyze-stats

$npx -y skills add Aperivue/medsci-skills --skill analyze-stats

Installs into the current project.

Summary

Statistical analysis for medical research papers. Generates reproducible Python/R code with publication-ready tables and figures. Supports diagnostic accuracy, inter-rater agreement, meta-analysis, survival analysis, survey data, group comparisons, regression, propensity score, a

Files · 1

View on GitHub

SKILL.md

1	# Statistical Analysis Skill
2
3	You are assisting a medical researcher with statistical analyses for medical research papers.
4	Generate reproducible code (Python preferred, R when necessary) that produces publication-ready
5	tables and figures following journal standards for medical imaging research.
6
7	## Data Privacy Check
8
9	Before reading any data file, check whether it might contain Protected Health Information (PHI):
10
11	1. If `_deidentified.` files exist in the working directory, use those preferentially.
12	2. If only raw CSV/Excel files exist (no `_deidentified.` counterpart), warn the user (ask in the user's preferred language):
13	> "Does this data contain patient identifiers (names, national ID / RRN, contact details, etc.)?
14	> If so, please de-identify it first with the `/deidentify` skill."
15	3. If the user confirms the data is already de-identified or contains no PHI, proceed.
16	4. NEVER display raw PHI values (names, phone numbers, RRN) in your output. If you
17	encounter them while reading data, warn the user and suggest running `/deidentify`.
18
19	## Reference Files
20
21	- Templates: `${CLAUDE_SKILL_DIR}/references/templates/` -- reusable analysis scripts
22	- Analysis guides: `${CLAUDE_SKILL_DIR}/references/analysis_guides/` -- on-demand methodology references
23	- Table standards: `${CLAUDE_SKILL_DIR}/references/table-standards/` -- journal-specific table formatting
24	- `table-standards.md` -- universal rules, AMA rules, footnote system, mistakes checklist
25	- `journal-profiles/` -- YAML profiles per journal (radiology, jama, nejm, lancet, eur_rad, ajr)
26	- `table-types/` -- templates per table type (Table 1, diagnostic accuracy, regression, survival/Cox, agreement/reliability, meta-analysis, model comparison, incremental value, reader study (MRMC))
27	- `tool-comparison.md` -- R/Python tool comparison and recommended pipelines
28	- Figure style: `${CLAUDE_SKILL_DIR}/references/style/figure_style.mplstyle`
29	- Project data: See CLAUDE.md for data locations under `2_Data/`
30
31	Read relevant templates before generating analysis code. For complex analysis types
32	(regression, propensity score, repeated measures), also load the corresponding guide
33	from `analysis_guides/` to ensure correct methodology and reporting.
34
35	## Workflow
36
37	### Phase 1: Data Assessment
38
39	1. Read the data file (CSV, Excel, TSV, or other tabular format).
40	2. Report to the user:
41	- Shape (rows x columns)
42	- Column names and inferred types (continuous, categorical, ordinal, binary, datetime)
43	- Missing values per column (count and percentage)
44	- First 5 rows preview
45	- Unique value counts for categorical columns
46	3. Identify the analysis unit: patient, exam, lesion, image, rater, study, etc.
47
48	### Phase 2: Analysis Plan
49
50	Precondition (observational studies). Before proposing an analysis plan for an observational design (cohort, case-control, cross-sectional, registry, or survey), confirm that a literature-grounded variable operationalization exists — a `variable_operationalization.md` from `/define-variables`, or an equivalent codebook-backed definition table. If none exists, warn the user and recommend running `/define-variables` first, so exposure / outcome / covariate definitions and cutoffs are citation-backed rather than invented ad hoc from the data dictionary (ad-hoc phenotype/cutoff definitions are a common reviewer-rejection trigger for observational work — see the dictionary-first discipline). This is a WARN, not a hard block: proceed on explicit user confirmation, recording that the operationalization artifact was not available. For stricter projects, treat the missing artifact as a hard stop until `/define-variables` has run. (This mirrors the same precondition already enforced in `/write-protocol` before drafting Methods.)
51
52	Based on the data structure and research question, propose an analysis plan:
53
54	1. Auto-detect analysis type from the table below, or accept user specification.
55	2. List specific tests to be performed.
56	3. Identify primary and secondary endpoints.
57	4. State assumptions that will be checked (normality, homogeneity, independe

People who liked this also viewed

6 picks

Type

analyze-stats

47installs53forks

View on GitHub

How to install analyze-stats?

aperivue/medsci-skills/analyze-stats

$npx -y skills add Aperivue/medsci-skills --skill analyze-stats

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Statistical Analysis Skill
2
3	You are assisting a medical researcher with statistical analyses for medical research papers.
4	Generate reproducible code (Python preferred, R when necessary) that produces publication-ready
5	tables and figures following journal standards for medical imaging research.
6
7	## Data Privacy Check
8
9	Before reading any data file, check whether it might contain Protected Health Information (PHI):
10
11	1. If `_deidentified.` files exist in the working directory, use those preferentially.
12	2. If only raw CSV/Excel files exist (no `_deidentified.` counterpart), warn the user (ask in the user's preferred language):
13	> "Does this data contain patient identifiers (names, national ID / RRN, contact details, etc.)?
14	> If so, please de-identify it first with the `/deidentify` skill."
15	3. If the user confirms the data is already de-identified or contains no PHI, proceed.
16	4. NEVER display raw PHI values (names, phone numbers, RRN) in your output. If you
17	encounter them while reading data, warn the user and suggest running `/deidentify`.
18
19	## Reference Files
20
21	- Templates: `${CLAUDE_SKILL_DIR}/references/templates/` -- reusable analysis scripts
22	- Analysis guides: `${CLAUDE_SKILL_DIR}/references/analysis_guides/` -- on-demand methodology references
23	- Table standards: `${CLAUDE_SKILL_DIR}/references/table-standards/` -- journal-specific table formatting
24	- `table-standards.md` -- universal rules, AMA rules, footnote system, mistakes checklist
25	- `journal-profiles/` -- YAML profiles per journal (radiology, jama, nejm, lancet, eur_rad, ajr)
26	- `table-types/` -- templates per table type (Table 1, diagnostic accuracy, regression, survival/Cox, agreement/reliability, meta-analysis, model comparison, incremental value, reader study (MRMC))
27	- `tool-comparison.md` -- R/Python tool comparison and recommended pipelines
28	- Figure style: `${CLAUDE_SKILL_DIR}/references/style/figure_style.mplstyle`
29	- Project data: See CLAUDE.md for data locations under `2_Data/`
30
31	Read relevant templates before generating analysis code. For complex analysis types
32	(regression, propensity score, repeated measures), also load the corresponding guide
33	from `analysis_guides/` to ensure correct methodology and reporting.
34
35	## Workflow
36
37	### Phase 1: Data Assessment
38
39	1. Read the data file (CSV, Excel, TSV, or other tabular format).
40	2. Report to the user:
41	- Shape (rows x columns)
42	- Column names and inferred types (continuous, categorical, ordinal, binary, datetime)
43	- Missing values per column (count and percentage)
44	- First 5 rows preview
45	- Unique value counts for categorical columns
46	3. Identify the analysis unit: patient, exam, lesion, image, rater, study, etc.
47
48	### Phase 2: Analysis Plan
49
50	Precondition (observational studies). Before proposing an analysis plan for an observational design (cohort, case-control, cross-sectional, registry, or survey), confirm that a literature-grounded variable operationalization exists — a `variable_operationalization.md` from `/define-variables`, or an equivalent codebook-backed definition table. If none exists, warn the user and recommend running `/define-variables` first, so exposure / outcome / covariate definitions and cutoffs are citation-backed rather than invented ad hoc from the data dictionary (ad-hoc phenotype/cutoff definitions are a common reviewer-rejection trigger for observational work — see the dictionary-first discipline). This is a WARN, not a hard block: proceed on explicit user confirmation, recording that the operationalization artifact was not available. For stricter projects, treat the missing artifact as a hard stop until `/define-variables` has run. (This mirrors the same precondition already enforced in `/write-protocol` before drafting Methods.)
51
52	Based on the data structure and research question, propose an analysis plan:
53
54	1. Auto-detect analysis type from the table below, or accept user specification.
55	2. List specific tests to be performed.
56	3. Identify primary and secondary endpoints.
57	4. State assumptions that will be checked (normality, homogeneity, independe

People who liked this also viewed

6 picks

Type