CtrlK
BlogDocsLog inGet started
Tessl Logo

nature-figure

Create, revise, audit, and export submission-grade scientific figures for Nature-family and other high-impact venues in Python (matplotlib/seaborn) or R (ggplot2/patchwork/ComplexHeatmap), including multi-panel plots, figures4papers-style work, and journal-ready SVG/PDF/TIFF outputs. Use for paper or scientific plots, manuscript data visualization, 论文配图、学术写作配图、科研绘图、科研作图、画图、作图、出图、论文图表、可视化. Define the conclusion, evidence logic, data integrity, template compatibility, export needs, and reviewer risks before plotting; honor or persist the Python/R backend choice. Also use the separate OpenRouter GPT Image 2 route for explicit AI-generated graphical abstracts, mechanism diagrams, concept schematics, 论文示意图、机制示意图、图形摘要; this route skips backend choice and treats outputs as drafts. Do not use for interactive dashboards, statistics-only analysis, data cleaning, literature review, code debugging, pure photo editing, or Illustrator/Figma-first infographics without manuscript-figure intent.

68

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Failed to scan

The risk profile of this skill

SKILL.md
Quality
Evals
Security

Quality

Content

72%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The router is well-sequenced with concrete scripts, explicit validation checkpoints, and a disciplined load-on-demand philosophy. However, the entire routing protocol hinges on a manifest.yaml and a static/ fragment layer that are not present in the bundle, which breaks the progressive-disclosure structure it advertises.

Suggestions

Ship the missing manifest.yaml and static/ files (contract.md, stance.md, fragments/backend/python.md and r.md) that steps 1–3 require, or rewrite the router to inline the minimal contract/stance so it is self-sufficient.

Move the inline NMI legend-word-count guidance fully into the nature-machine-intelligence.md reference and reference it by pointer, keeping the router to the routing rule.

Verify the ../nature-shared/journal-formats/nature-machine-intelligence.md path resolves from the skill directory, since a broken relative path silently disables the NMI branch.

DimensionReasoningScore

Conciseness

The router is lean for its scope and repeatedly tells Claude to load fragments from disk rather than reproducing them, but it still re-explains journal-specific rules inline (e.g., the full NMI word-count advisory and 'count the whole legend, not each panel') that belong in the loaded reference, leaving minor over-explanation to trim.

4 / 5

Actionability

Concrete, executable commands are given throughout (`scripts/nature_figure_backend.py set python`, `scripts/nature_figure_backend.py get`, `scripts/validate_figure.py`, `scripts/audit_pdf_text.py`) with exact file paths to Read; it stops just short of copy-paste-ready examples because the actual code/quick-start lives in the on-demand fragments by design.

4 / 5

Workflow Clarity

A clearly numbered routing protocol (0–5) sequences every invocation with a blocking backend gate, an ordered build step, and an explicit delivery preflight with validation checkpoints (validate_figure.py → audit_pdf_text.py → panel-by-panel audit), plus error-recovery framing ('If no saved preference exists, ask… and stop').

5 / 5

Progressive Disclosure

Although the body signals many on-demand references, several core paths it instructs Claude to load are missing from the bundle (manifest.yaml, static/core/contract.md, static/core/stance.md, static/fragments/backend/python.md and r.md, ../nature-shared/journal-formats/nature-machine-intelligence.md) — the central routing mechanism is broken because the static layer and manifest the whole protocol depends on are absent, so the disclosure structure does not actually resolve.

2 / 5

Total

15

/

20

Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A thorough, third-person description with concrete actions, dense bilingual trigger terms, explicit use-when guidance, and clear negative boundaries. Its only weakness is verbosity — it packs two full skill routes and long keyword lists into one block, which borders on padding.

Suggestions

Tighten the description by moving the AI-graphical-abstract route's trigger keyword list into the body; the description could state the route once and keep the Chinese synonym dump shorter.

Consider trimming the negative-boundary list to the highest-contrast exclusions to improve signal-to-noise without losing distinctiveness.

DimensionReasoningScore

Specificity

Lists multiple concrete actions ('Create, revise, audit, and export submission-grade scientific figures'), names backends (matplotlib/seaborn, ggplot2/patchwork/ComplexHeatmap) and output formats (SVG/PDF/TIFF), and covers multi-panel plots, figures4papers, and a separate AI schematic route — comprehensive coverage.

5 / 5

Completeness

Explicitly answers 'what' (figure creation/revision/audit/export across Python and R) and 'when' ('Use for paper or scientific plots, manuscript data visualization…'), plus a separate 'Also use…' clause for the AI route and a 'Do not use for…' boundary — both what and when are concrete and explicit.

5 / 5

Trigger Term Quality

Dense natural trigger coverage including 'paper or scientific plots', 'manuscript data visualization', plus extensive Chinese synonyms (论文配图, 科研绘图, 画图, 出图, 论文图表) and AI-schematic terms (论文示意图, 机制示意图, 图形摘要), matching how multilingual users actually phrase requests.

5 / 5

Distinctiveness Conflict Risk

The Nature/scientific-figure niche with backend split and explicit 'Do not use for…' exclusions (dashboards, statistics-only, data cleaning, debugging, photo editing, infographics) is largely distinct; minor overlap risk with general plotting or sibling nature-* skills keeps it just below a clear-niche 5.

4 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 missing

Warning

Total

15

/

16

Passed

Repository
Yuan1z0825/nature-skills
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.