Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Light0305/Light-skillsnpx agentmods add skills/light0305/light-skills/light-figureWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/light0305/light-skills/light-figure)<a href="https://agentmods.dev/skills/light0305/light-skills/light-figure"><img src="https://agentmods.dev/badge/skills/light0305/light-skills/light-figure/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/light0305/light-skills/light-figure"><img src="https://agentmods.dev/badge/skills/light0305/light-skills/light-figure.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00514 | $0.08189 |
| Opus 5 | $0.00257 | $0.04095 |
| Sonnet 5 | $0.00103 | $0.01638 |
| Haiku 4.5 | $0.00051 | $0.00819 |
Grade A, and why
light-figure scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 228 lines — stays where its author put it; the contents beside it link to each section on GitHub.
图表（figure）—— 科研主线 stage 9 · 视觉诚实(critical) + 图服务论点 + 程序化生成
你是 Light 科研流水线的 DAG 第 9 节点（v1 的 figure-planning + figure-drawing 合并）。任务不是「把数据画成好看的图」， 是让每张图都为论点服务、达出版级、且绝不撒谎：规划（图服务哪条 claim + 集合预算 + 反冗余）→ 绘制（出版级 + 色盲安全 + 误差棒）→ 诚实门（截/双轴/把不显著当主图）→ 渲染后真看一眼。守住两条红线：视觉不诚实 = critical、论文数据图 绝不 AI 生图。
一句话定位：把「一屋子院士看你的图时真正死磕的」——这图诚实吗（y 轴有没有偷偷截断放大差异、双 y 轴有没有制造 伪相关、用没用 jet/rainbow 误导）、误差棒标了吗（类型 SD/SEM/CI + n）、黑白/色盲能辨吗、这图支撑哪条 claim、 删了缺什么（还是把不显著结果硬当主图）、一组图超没超 venue 预算、有没有冗余 panel——落成确定性机读门 （视觉不诚实=critical）+ 集合预算/反冗余自检 + render-then-look 渲染回看 + 把不显著当主图→回炉。 深度对标真相源 =
docs/competitors/figure.md（Round 2 R1：10 真·同类绘图 skill 实搜+读码， scipilot 530★/davila7 28.2K★ 等头部 + 机制对标 + 超越点 + 诚实边界；诚实校正：截轴/双轴/render-then-look 是同类共识， Light 增量=确定性机读门 + 绑证据档 misrepresent_evidence + 9→7 回炉，非"想到截轴"）。谁产 findings、谁是 critical 门（诚实分工）：本技能产视觉诚实 findings（producer=figure，
visual_honesty_gate.py六 gate）——visual_honesty（截 y 轴/双 y 轴伪相关/3D 透视）+misrepresent_evidence（把不显著当主图/图文不一致）= critical，被run_checkpoint --stage 9聚合 → critical fail exit 1；error_bars/colormap/display_budget/panel_redundancy= warn 不阻断 DAG（spec §4.2 口径）。与
_shared/visual_qa的分工（别重造几何/对比度引擎）：_shared/visual_qa已建几何检测（标签重叠/溢出/对齐）+ WCAG 对比度门（两档）+ render-then-look 协议（批 0 地基）。figure 只编排消费——经figure_visual_qa.py抽 matplotlib 元素的 AABB 喂给它的几何引擎，不重写 detect_geometry_issues（同 paper-writing 消费 evidence_contract 不重造 lint 的范式）。 figure 的真增量 = visual_qa 没有的：截/双轴/rainbow 的诚实判据（figure_integrity_lint）+ 集合级预算/反冗余 （audit_figure_set）+ 图↔claim 绑定（消费 evidence_strength.json）。与 result-analysis / paper-writing 的分工：result-analysis 定证据档（emit
evidence_strength.json+ result card）； paper-writing 规划哪些 claim 要图支撑，并给出 guardrail claim_impact/limitations；figure 据证据档定图注统计标注 + 卡「把不显著当主图」，并要求 guardrail WARN 的限制进入 caption。三者共用 result card / claim plan 这条交接链。特殊位置（回炉发起方）：规划/渲染时发现某主图绑的 claim 不显著（grade=none）、或图与正文 claim 不一致 → findings 带「不显著/图文不一致/证据/主图」信号 → 总控
reroute --stage 9建议 9→7 回 result-analysis 核「该图对应的证据强度」。 截/双轴这类画法不诚实不回炉——issue 不带信号 → reroute 给 manual = 本阶段重画修（同 result-analysis p-hacking→manual 的诚实落点）。回炉是决策点，停下问用户。是横切常驻吗？ 否。这是按需
/调用的主线节点；file-reading / memory-pm / project-structure / consistency / research-ethics 全程横切常驻，本技能不重复它们。
What ships with it
23 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/nature.mplstyle 1.6 KB
- assets/publication.mplstyle 1.7 KB
- assets/science.mplstyle 1.7 KB
- examples/bad_figure_example.py 1.7 KB runs code
- examples/example_framework_render.py 3.6 KB runs code
- examples/example_framework.dot 1.3 KB
- examples/example_matplotlib_multipanel.py 4.7 KB runs code
- examples/example_seaborn_stats.py 3.4 KB runs code
- examples/worked_example.md 6.8 KB
- references/figure_integrity.md 5.8 KB
- scripts/audit_figure_set.py 17 KB runs code
- scripts/color_palettes.py 7.2 KB runs code
- scripts/figure_contract.py 26 KB runs code
- scripts/figure_export.py 30 KB runs code
- scripts/figure_integrity_lint.py 8.8 KB runs code
- scripts/figure_visual_qa.py 14 KB runs code
- scripts/r_ggplot.py 42 KB runs code
- scripts/recommend_chart.py 13 KB runs code
- scripts/validate_plan_card.py 12 KB runs code
- scripts/visual_honesty_gate.py 34 KB runs code
- templates/figure_plan_card.md 5.3 KB
- templates/figure-delivery.example.json 3.0 KB
- templates/table_plan_card.md 4.4 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 228 lines · 514 tokens per session scan A 2bb0ae7cbbc8
light-figure is a skill published in the GitHub repository Light0305/Light-skills (610 stars, last pushed 2mo ago), licensed MIT. It adds 514 tokens to every session and 8,189 once invoked, about $0.0026 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
anti-defensive-writing-en
Stops defensive writing across the entire paper lifecycle — writing, revising, cutting, and organizing experiments. Treats the paper as a press conference, not a project summary, lab log, or self-audit: identify the single most publishable strength of the work and build the most favorable, complete, and persuasive…
anti-defensive-writing
A Chinese-language writing guide for presenting a research paper around its strongest supported contribution. It treats the paper as a focused academic presentation rather than a project diary or complete lab record.
research-writing
A collection of 30 prompt templates for writing and reviewing scientific papers. It covers tasks such as translating, editing, summarizing research, writing sections, creating figure captions, and preparing reviewer replies.
ml-paper-writing
Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, reviewer guidelines, and citation verification workflows.
ml-paper-writing
Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, conducting literature reviews, finding related work, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, citation verification workflows, and paper…
academic-citation
Search, verify, and map citations for CS/AI/ML papers. Produces VERIFIED/UNVERIFIED reference lists with Citation-to-Claim maps and Exemplar Sets. Use when: finding references for a paper section, verifying citation accuracy, building exemplar sets for introduction/related work learning, checking if existing citations…