cn-thesis-data-figures

CN Thesis Data Figures

This skill uses a router plus manifest architecture. Do not apply the data/figure workflow from memory.

Routing Protocol

1. Read manifest.yaml.
2. Read every path listed in always_load.
3. Detect the axis values from the user request:

• degree_level: master, doctoral, professional, or undergraduate.
• discipline: life-sciences, computer-science, humanities-social-sciences, engineering, or generic.
• thesis_type: empirical-experimental, mechanism-hypothesis, method-protocol, engineering-system, review-synthesis, or case-text-analysis.
• language: zh, en, or bilingual.
• analysis_mode: data-audit, statistics-plan, figure-design, reproducibility-check, or result-interpretation.

1. Read only the fragments mapped to the detected values.
2. State the detected values in one short line before the main answer.
3. Run static/core/workflow.md end to end.

Boundaries

• Define the figure/data contract before giving analysis or chart advice.
• When the user needs a thesis-wide figure/table inventory for later drafting, use or update output/thesis-context/figure-table-list.md with thesis location, bilingual caption/title, evidence shown, supported claim, and missing facts.
• Do not invent data, sample size, statistical significance, or image content.
• Apply the thesis figure style specification for typography, axes, lines, color, grayscale printing, and multi-panel layout when figures are designed or reviewed.
• Keep Word/docx thesis insertion needs in view unless the user chooses another target.