figure-table-audit

Name: figure-table-audit
Author: scdenney

34stars0forks

How to install figure-table-audit?

scdenney/open-science-skills/figure-table-audit

$npx -y skills add scdenney/open-science-skills --skill figure-table-audit

Installs into the current project.

Summary

Audit figures, tables, captions, cross-references, and statistical notes.

Files · 1

View on GitHub

SKILL.md

1	# Figure and Table Auditor
2
3	## Heritage and scope
4
5	This is an original Open Science Skills workflow for manuscript QA. It remixes general figure/table and citation-compliance ideas from Cheng-I Wu's Academic Research Skills for Claude Code (CC BY-NC 4.0), but is rewritten for open-science social-science manuscripts. It is not a visual hallucination engine: when a claim requires reading plotted values from an image, prefer source data or mark the issue as needing author verification.
6
7	This is the end-stage auditor. For figure design and production guidance during drafting, use the `figures` skill; for table design, use the `tables` skill. Run `figure-table-audit` once the figure and table set is stable and you are preparing for submission.
8
9	## Instructions
10
11	### 1. Orient before auditing
12
13	Identify:
14
15	- Manuscript source and compiled output, if any.
16	- Figure directory, table files, appendix/SI files, and build command.
17	- Target journal or style guide.
18	- Whether figures are generated from code, manually edited, or exported from software.
19	- Whether source data for figures/tables are available.
20
21	If only a PDF is available, state that cross-reference and value checks are lower confidence.
22
23	### 2. Inventory figures and tables
24
25	Build an inventory with:
26
27	- Figure/table number or label.
28	- File path or source location.
29	- Caption/title.
30	- First in-text callout.
31	- Appendix/SI location if applicable.
32	- Producing script or data source, if visible.
33
34	Check:
35
36	- Every in-text callout resolves to an existing figure/table.
37	- Every figure/table is called out in text before or near appearance.
38	- Numbering is gap-free and not duplicated.
39	- Main-text and SI labels do not collide.
40	- LaTeX/Markdown references compile without unresolved labels.
41
42	### 3. Check text-to-evidence consistency
43
44	For each figure/table used to support a substantive claim:
45
46	- Match the in-text claim to the exact row, column, panel, model, or plotted quantity.
47	- Verify direction, magnitude, uncertainty, subgroup, and denominator.
48	- Check whether text overstates non-significant or imprecise estimates.
49	- Check whether figure/table notes disclose model specification, N, weights, fixed effects, clustering, and missing-data handling.
50	- For experimental papers, check ITT vs per-protocol labeling, attrition-by-arm, baseline balance, and CONSORT/sample-flow consistency.
51
52	Do not infer exact values by eyeballing a plot unless the figure encodes labeled values. If source data are unavailable, write `VISUAL READ ONLY - AUTHOR VERIFY`.
53
54	### 4. Audit captions and notes
55
56	Captions and table notes should let a reader understand the evidence without hunting:
57
58	- Figure caption states what is plotted, units, sample, uncertainty interval, and panel meanings.
59	- Table title states the estimand or model family, not only "Results."
60	- Table notes define dependent variable, treatment/condition coding, omitted categories, controls, fixed effects, clustering, weights, and significance markers.
61	- All abbreviations are defined on first use.
62	- All transformations, scales, and index directions are explicit.
63	- Any data exclusions or subgroup restrictions are named.
64
65	For conjoint, list-experiment, topic-modeling, text-classification, and OCR studies, invoke or recommend the relevant sibling skill when table/figure interpretation depends on method-specific standards.
66
67	### 5. Check accessibility and production quality
68
69	Flag:
70
71	- Color palettes not interpretable in grayscale or by color-blind readers.
72	- Low-resolution or rasterized text in vector outputs.
73	- Axis labels too small or missing units.
74	- Legends that obscure data or use ambiguous labels.
75	- Inconsistent decimal precision.
76	- Overcrowded tables that should move to SI.
77	- Missing alt text when the target venue requires it.
78	- Non-reproducible manual edits that are not documented.
79
80	### 6. Audit SI and replication linkage
81
82	Check whether:
83
84	- Main-text claims that point to SI land on the correct appendix/table/figure.
85	- SI numbering and captions are internally consistent.
86	- Each main figure/table can be traced to a script or documented manual step.
87	- Replication README maps outputs to scripts and data files.
88	- Figure/table source data are archived or clearly restricted.
89
90	## Output
91
92	Produce a `Figure and Table Audit Report`:
93
94	```
95	# Figure and Table Audit Report
96
97	Scope:
98	Inputs checked:
99	Build/source status:
100	Summary: <N blocking, N recommended, N minor, N author-verification>
101
102	## Inventory
103	\| ID \| Path/location \| Caption/title \| First callout \| Source/script \|
104
105	## Blocking Issues
106	\| Location \| Figure/table \| Issue \| Evidence \| Fix \|
107
108	## Recommended Fixes
109	\| Location \| Figure/table \| Issue \| Fix \|
110
111	## Minor / Production Issues
112	\| Figure/table \| Issue \| Fix \|
113
114	## Author Verification Needed
115	\| Figure/table \| Why verification is needed \|
116
117	## Readiness Checklist
118	\| Dimension \| PASS/FAIL/PARTIAL/NA \| Notes \|
119	```
120
121	Severity:
122
123	- Blocking: mis

People who liked this also viewed

6 picks

Type

find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 20262.6M27k
grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026633k182k
vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026573k29k
improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026522k182k
setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026454k182k
triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026424k182k

figure-table-audit

34stars0forks

View on GitHub

How to install figure-table-audit?

scdenney/open-science-skills/figure-table-audit

$npx -y skills add scdenney/open-science-skills --skill figure-table-audit

Installs into the current project.

Summary

Audit figures, tables, captions, cross-references, and statistical notes.

Files · 1

View on GitHub

SKILL.md

1	# Figure and Table Auditor
2
3	## Heritage and scope
4
5	This is an original Open Science Skills workflow for manuscript QA. It remixes general figure/table and citation-compliance ideas from Cheng-I Wu's Academic Research Skills for Claude Code (CC BY-NC 4.0), but is rewritten for open-science social-science manuscripts. It is not a visual hallucination engine: when a claim requires reading plotted values from an image, prefer source data or mark the issue as needing author verification.
6
7	This is the end-stage auditor. For figure design and production guidance during drafting, use the `figures` skill; for table design, use the `tables` skill. Run `figure-table-audit` once the figure and table set is stable and you are preparing for submission.
8
9	## Instructions
10
11	### 1. Orient before auditing
12
13	Identify:
14
15	- Manuscript source and compiled output, if any.
16	- Figure directory, table files, appendix/SI files, and build command.
17	- Target journal or style guide.
18	- Whether figures are generated from code, manually edited, or exported from software.
19	- Whether source data for figures/tables are available.
20
21	If only a PDF is available, state that cross-reference and value checks are lower confidence.
22
23	### 2. Inventory figures and tables
24
25	Build an inventory with:
26
27	- Figure/table number or label.
28	- File path or source location.
29	- Caption/title.
30	- First in-text callout.
31	- Appendix/SI location if applicable.
32	- Producing script or data source, if visible.
33
34	Check:
35
36	- Every in-text callout resolves to an existing figure/table.
37	- Every figure/table is called out in text before or near appearance.
38	- Numbering is gap-free and not duplicated.
39	- Main-text and SI labels do not collide.
40	- LaTeX/Markdown references compile without unresolved labels.
41
42	### 3. Check text-to-evidence consistency
43
44	For each figure/table used to support a substantive claim:
45
46	- Match the in-text claim to the exact row, column, panel, model, or plotted quantity.
47	- Verify direction, magnitude, uncertainty, subgroup, and denominator.
48	- Check whether text overstates non-significant or imprecise estimates.
49	- Check whether figure/table notes disclose model specification, N, weights, fixed effects, clustering, and missing-data handling.
50	- For experimental papers, check ITT vs per-protocol labeling, attrition-by-arm, baseline balance, and CONSORT/sample-flow consistency.
51
52	Do not infer exact values by eyeballing a plot unless the figure encodes labeled values. If source data are unavailable, write `VISUAL READ ONLY - AUTHOR VERIFY`.
53
54	### 4. Audit captions and notes
55
56	Captions and table notes should let a reader understand the evidence without hunting:
57
58	- Figure caption states what is plotted, units, sample, uncertainty interval, and panel meanings.
59	- Table title states the estimand or model family, not only "Results."
60	- Table notes define dependent variable, treatment/condition coding, omitted categories, controls, fixed effects, clustering, weights, and significance markers.
61	- All abbreviations are defined on first use.
62	- All transformations, scales, and index directions are explicit.
63	- Any data exclusions or subgroup restrictions are named.
64
65	For conjoint, list-experiment, topic-modeling, text-classification, and OCR studies, invoke or recommend the relevant sibling skill when table/figure interpretation depends on method-specific standards.
66
67	### 5. Check accessibility and production quality
68
69	Flag:
70
71	- Color palettes not interpretable in grayscale or by color-blind readers.
72	- Low-resolution or rasterized text in vector outputs.
73	- Axis labels too small or missing units.
74	- Legends that obscure data or use ambiguous labels.
75	- Inconsistent decimal precision.
76	- Overcrowded tables that should move to SI.
77	- Missing alt text when the target venue requires it.
78	- Non-reproducible manual edits that are not documented.
79
80	### 6. Audit SI and replication linkage
81
82	Check whether:
83
84	- Main-text claims that point to SI land on the correct appendix/table/figure.
85	- SI numbering and captions are internally consistent.
86	- Each main figure/table can be traced to a script or documented manual step.
87	- Replication README maps outputs to scripts and data files.
88	- Figure/table source data are archived or clearly restricted.
89
90	## Output
91
92	Produce a `Figure and Table Audit Report`:
93
94	```
95	# Figure and Table Audit Report
96
97	Scope:
98	Inputs checked:
99	Build/source status:
100	Summary: <N blocking, N recommended, N minor, N author-verification>
101
102	## Inventory
103	\| ID \| Path/location \| Caption/title \| First callout \| Source/script \|
104
105	## Blocking Issues
106	\| Location \| Figure/table \| Issue \| Evidence \| Fix \|
107
108	## Recommended Fixes
109	\| Location \| Figure/table \| Issue \| Fix \|
110
111	## Minor / Production Issues
112	\| Figure/table \| Issue \| Fix \|
113
114	## Author Verification Needed
115	\| Figure/table \| Why verification is needed \|
116
117	## Readiness Checklist
118	\| Dimension \| PASS/FAIL/PARTIAL/NA \| Notes \|
119	```
120
121	Severity:
122
123	- Blocking: mis

People who liked this also viewed

6 picks

Type