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-paper-writingWrote 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-paper-writing)<a href="https://agentmods.dev/skills/light0305/light-skills/light-paper-writing"><img src="https://agentmods.dev/badge/skills/light0305/light-skills/light-paper-writing/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-paper-writing"><img src="https://agentmods.dev/badge/skills/light0305/light-skills/light-paper-writing.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 | $0.00435 | $0.10223 |
| Opus 5 | $0.00217 | $0.05111 |
| Sonnet 5 | $0.00087 | $0.02045 |
| Haiku 4.5 | $0.00044 | $0.01022 |
Grade A, and why
light-paper-writing 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 12d 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 — 315 lines — stays where its author put it; the contents beside it link to each section on GitHub.
论文写作(paper-writing)—— 科研主线 stage 8 · claim 必有证据(诚信门) + 审稿人视角循环
你是 Light 科研流水线的 DAG 第 8 节点。任务不是「把字凑成一篇论文」,是围绕一个问题组织全文——怎么让 审稿人相信这工作值得发表:初稿 → 自己当审稿人挑一遍 → 循环打磨。守住一条红线:每个 claim 都有证据、措辞强度 匹配证据强度、绝不过度宣称。claim 无证据 = critical 诚信门;过度宣称、贡献三处不一致 = warn。
一句话定位:把「一屋子院士在审你论文时真正死磕的」——每个 claim 有证据吗(摘要里说的,正文撑得住吗)、 措辞配不配证据(强证据强措辞、弱证据 hedge、不显著只能说「未见显著差异」、绝不 spin)、贡献三处一致吗 (摘要/引言/结论说的是不是同样几条、数字一致)、引言把痛点→不足→洞察→贡献讲清了吗、审稿人会从哪挑—— 落成确定性机读门(claim 无证据=critical)+ 自检 findings + 审稿人视角循环。 深度对标真相源 =
docs/competitors/paper-writing.md(8 个真同类 skill + 机制锚 + 超越点 + 诚实边界)。谁产 findings、谁是 critical 门(诚实分工):本技能产 claim 必有证据/过度宣称/result-card guardrail findings(producer=paper-writing,
claim_evidence_gate.py五 gate)——claim_evidence(草稿强断言未登记、未绑定自己的 evidence claim_id、绑定不存在或只绑定 none → critical 诚信门)被run_checkpoint --stage 8聚合 → critical fail exit 1;overclaim(措辞强于证据档)、contribution_consistency(贡献三处漂移)= warn 不阻断 DAG(spec §4.2 口径);result_card_guardrail会把缺 result-card、未 ready、guardrail blocking 或 WARN 未限制的实证 claim 作为 critical。与 research-ethics 的分工(evidence_contract 是桥,别两套重造 lint_wording):paper-writing = 写作时自检 (claim 无证据 critical + 过度宣称 warn,写的时候自己先 lint);research-ethics = 交付前横切硬红线(
claim_evidence_bind的conclusion_overclaim,措辞超档 = critical)。两者共用同一个_shared/evidence_contract.lint_wording——单一措辞 引擎、两消费方、两语境(写作自检 vs 交付红线)、两严重度(warn vs critical)。不重造措辞档。与 consistency 的分工:paper-writing = 单稿内贡献三处对齐自检(abstract/intro/conclusion,warn);C2 consistency = 跨材料/跨阶段术语·指标·创新点一致性常驻复核(论文↔slides↔lit)。不重造一致性引擎。
特殊位置(回炉发起方):写作时发现某 claim 无实验支撑 → findings 带「claim/证据/支撑」信号 → 总控
reroute --stage 8建议 8→7 回 result-analysis 补证据(默认);若该结论实验根本没产出 → 改 8→6 回 experiment-coding 补实验。 这是决策点,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。真实作者工作流:先读
paper-writing-resource-map.md。它把 venue/claim plan、证据收齐、 section contract、reverse outline、自审、机读门、回炉与下游交接串成六步闭环;不是资源网址罗列。
What ships with it
21 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.
- examples/claim_evidence_walkthrough.md 3.3 KB
- paper-writing-resource-map.md 9.4 KB
- references/argument_review.md 4.9 KB
- references/claim_argument_plan.md 5.5 KB
- references/guideline_map.md 4.3 KB
- references/integrity_gate.md 4.4 KB
- references/mandatory_inclusions.md 4.0 KB
- references/self_review_checklist.md 3.3 KB
- scripts/argument_contract.py 19 KB runs code
- scripts/claim_binding.py 45 KB runs code
- scripts/claim_evidence_gate.py 34 KB runs code
- scripts/contribution_consistency.py 13 KB runs code
- scripts/draft_lint.py 21 KB runs code
- scripts/mechanical_check.py 24 KB runs code
- scripts/polish.py 13 KB runs code
- scripts/style_fingerprint.py 8.7 KB runs code
- templates/01_imrad.md 2.3 KB
- templates/06_conference.md 1.7 KB
- templates/argument-outline.example.json 1.3 KB
- templates/claim_argument_plan.json 2.3 KB
- templates/claim_passport.md 4.1 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.
- 12d ago First seen · 315 lines · 435 tokens per session scan A 584adfa21fb1
light-paper-writing is a skill published in the GitHub repository Light0305/Light-skills (620 stars, last pushed 2mo ago), licensed MIT. It adds 435 tokens to every session and 10,223 once invoked, about $0.0022 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…