Use when building or auditing multi-panel mechanistic figures and tables for a Cancer Cell (Cell Press) manuscript — quantification of images, panel logic, image integrity, and legends. It shapes display items; it does not choose statistical tests or write the abstract.
SkillScore86Maintained
Automatic
◆Research pack
Last verified 8/31/2026
What it changes
Passed 19 of 22 test cases with the skill, and 17 of 22 without it.
Without skill77%
With skill86%
Checks it passed that the model alone missed
Includes the header 【Legends】 in the output.
Specifies cc-structured-abstract under 【Next step】.
Recommends cc-structured-abstract as the next step.
and 7 more in the full report
4 fixed15 passed either way1 still failing2 broke
Pass-rate lift
+9 pp
Tokens
+41%
Agent turns
0%
Response time
-11%
Model gemini-3.6-flashMeasured 8/31/2026v119 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 4
Prompt
I'm preparing our manuscript draft's figure set for Cancer Cell submission. Please review our main figures (Figures 1-4) containing Western blots, IHC slides, tumor growth curves, and flow cytometry plots. Can you audit them and outline: (1) the core message conveyed by each figure, (2) any representative panels that lack accompanying quantitative data across replicates, (3) potential image integrity concerns such as raw blot availability or contrast adjustments, (4) whether the figure legends adequately report sample sizes, statistical tests, error bar definitions, and scale bars, (5) whether required in vivo or human validation panels are present, and (6) the logical next step in the manuscript pipeline?
What was checked, without and with the skill
Checked
Without skill
With skill
Includes the header 【Per-figure message】 in the output.
Missed
Passed
Includes the header 【Representative-without-quantification】 in the output.
Missed
Passed
Specifies cc-structured-abstract under 【Next step】.
Missed
Passed
case-01 · gemini-3.6-flash · 8/31/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md117 linessingle file
Figures & Tables (cc-figures-tables)
When to trigger
Representative images (blots, IHC, micrographs) shown without quantification
A multi-panel figure has no clear narrative logic
Worried about image-integrity standards (splicing, over-processing)
Building a graphical abstract or supplemental display items
Cancer Cell figure logic
Each main figure should advance one step of the mechanistic argument and be readable on its own:
Panels flow left-to-right / top-to-bottom as a mini-narrative (perturbation → readout → mechanism → consequence).
Mechanistic figures typically combine schematic + functional data + in vivo and/or human validation.
A figure usually carries discovery, validation, and a control panel together.
Keep one main message per figure; push orthogonal corroboration to supplement.
Quantify everything representative
The signature Cancer Cell requirement: a representative image must be paired with quantification across biological replicates.
Representative image
Required quantification
Western blot
Densitometry across n independent blots (normalized to loading)
IHC / IF micrograph
Scored / counted across multiple fields and tumors (e.g., QuPath)
Flow plot
Summary bar/dot plot of % or MFI across replicates
Tumor photo
Tumor volume/weight curves with n, stats
Migration/invasion image
Quantified across fields/replicates
Single-cell UMAP
Proportions / DE stats, not just a colored embedding
Show data points (superplots) for small n; never imply quantity from one picture.
Image integrity (Cell Press enforces)
Keep uncropped, unprocessed raw blots/gels; include full-length blots in supplement with MW markers.
No splicing of lanes without a clear dividing line and disclosure.
Adjustments (brightness/contrast) must be linear and applied to the entire image, including controls.
Do not duplicate, mirror, or reuse panels across figures.
Disclose any grouping of images from different parts of the same gel.