Borrowing it
Nothing to install: this file belongs to dengxu11111/dual-ai-paper-coach. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/dengxu11111/dual-ai-paper-coach/main/.claude/commands/draft.mdgit clone --depth 1 https://github.com/dengxu11111/dual-ai-paper-coachWrote 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/commands/dengxu11111/dual-ai-paper-coach/draft)<a href="https://agentmods.dev/commands/dengxu11111/dual-ai-paper-coach/draft"><img src="https://agentmods.dev/badge/commands/dengxu11111/dual-ai-paper-coach/draft/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/commands/dengxu11111/dual-ai-paper-coach/draft"><img src="https://agentmods.dev/badge/commands/dengxu11111/dual-ai-paper-coach/draft.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00016 | $0.01179 |
| Opus 5 | $0.00008 | $0.00589 |
| Sonnet 5 | $0.00003 | $0.00236 |
| Haiku 4.5 | $0.00002 | $0.00118 |
Grade A, and why
draft 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 11d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/draft — 论文初稿生成
你是这个项目的主写作 agent。读 prompts/writer_system.md 作为系统准则,然后按 prompts/draft_paper.md 的模板产出一份完整的论文 markdown。
⚠️ 写作质量标杆:Science / Nature——即使实际投 GRL/ACP/JGR,本仓库也按 S/N 尺子收紧(≤ 4500 词 main text、≤ 5 displays、≤ 50 refs、Nature Reporting Summary 统计完整度、abstract 第一句必须给 broad significance)。具体细则见 prompts/writer_system.md 的 "Science / Nature Quality Standards" 段。
输入
- 系统准则:
prompts/writer_system.md(hard rules:不编数据、不编引用、保守措辞) - 任务模板:
prompts/draft_paper.md(论文 8 段结构:Title / Abstract / Intro / Methods / Results / Discussion / Conclusions / References) - 研究主题:用户输入的
$ARGUMENTS(一句话研究问题,例如 "全球气溶胶 2010-2024 时空演变") - 数据:
my-paper/data/analysis_results.json(如果存在 → 用真实数字;不存在 → 标[NUMBER_NEEDED: ...]) - 可选:
my-paper/CLAUDE.md里的领域上下文
步骤
- 读
prompts/writer_system.md,把它的内容当作你写作的 hard rules(特别是"never invent"和"conservative claims") - 读
prompts/draft_paper.md模板,把里面的{topic}替换成$ARGUMENTS - 检查
my-paper/data/analysis_results.json是否存在- 存在:把 JSON 内容附到 prompt 末尾,作为
{data_section} - 不存在:在 prompt 里说明"数据缺失,所有数字标 [NUMBER_NEEDED]"
- 存在:把 JSON 内容附到 prompt 末尾,作为
- 写一份完整论文 markdown(Title → Abstract → Intro → Methods → Results → Discussion → Conclusions → References)
- 把结果写到
my-paper/draft.md - 同时把这次运行的元数据存到
my-paper/runs/<ISO时间戳>/draft_meta.json:- 用了哪个版本的 prompts
- analysis_results.json 是否存在
- 论文字数、段落数
输出
my-paper/draft.md— 论文初稿my-paper/runs/<ts>/draft_meta.json— 元数据- 在终端用 markdown 预览前 30 行让用户看到
提醒用户的下一步
输出末尾打印:
初稿已写到
my-paper/draft.md。下一步:
- 运行
/codex-review让 Codex 独立审稿- 第一稿典型分数 4-5/10(S/N 标杆下偏低是正常的——Codex 按 Nature 审稿人尺度量你)
- 这比"普通期刊审稿"低 1-2 分是设计上的特性,不是 bug
红线
- ❌ 不要把 prompts/reviewer_system.md 读进来(那是 Codex 的,跟你写作无关,混进来会污染独立性)
- ❌ 不要编数字、引用、作者名
- ❌ 默认保守措辞:"is consistent with" 而不是 "proves"
失败排查
| 症状 | 原因 | 解决 |
|---|---|---|
| 报 "prompts/writer_system.md not found" | 工作目录不对 | 确认 cwd 是项目根(含 prompts/ 目录),不是 my-paper/ 子目录 |
draft.md 里大量 [NUMBER_NEEDED] |
analysis_results.json 不存在或字段缺 | 回到数据准备步骤补全 JSON:study_period / main_finding / seasonal_findings 三段必须有 |
| 输出全是中文 | writer_system.md 的 OUTPUT LANGUAGE 被改了 | 检查 prompts/writer_system.md 第一行是否还是 OUTPUT LANGUAGE: English |
| 字数 > 5500 大幅超 | draft_paper.md 模板被改 | 检查 prompts/draft_paper.md Constraints 段是否还有 ≤ 4500 words(S/N 标杆收紧版) |
| Abstract 第一句不是 broad significance | writer_system.md 的 S/N 段没被遵守 | 重读 prompts/writer_system.md "Significance framing" 段,把 Abstract sentence 1 改成 multidisciplinary 读者能听懂的层级 |
| Results 没有 CI / n / 检验名 | 不符合 Nature Reporting Summary | 让 Claude 把每个数字补齐 5 元组:effect size + 95% CI + n + 检验名 + 精确 p;缺哪个标 [NUMBER_NEEDED: ...] |
| Claude 给出真 DOI | 这就是 hallucination | 立刻去 doi.org 验证;任何引用 DOI 都不能信任 Claude 的输出 |
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.
- 11d ago First seen · 65 lines · 16 tokens per session scan A 9fc3f7a05332
draft is a command published in the GitHub repository dengxu11111/dual-ai-paper-coach (12 stars, last pushed 3mo ago), licensed MIT. It adds 16 tokens to every session and 1,179 once invoked, about $0.0001 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 commands, from other repositories
verify-math
Verify a self-authored mathematical result end to end by routing claims across adversarial review, numerical falsification, symbolic or CAS checks, and Lean, then aggregating one report. Use when a theorem, proposition, conjecture, or paper-wide mathematical argument needs the appropriate combination of verification…
replication-package
Scaffold or audit a social-science replication package at a target directory, and audit the manuscript and its archived research objects against FAIR principles.
diff
Quantitative volume comparison between a CadQuery model and a reference STEP file.
simulation-calibrator
Test and refine simulation accuracy with validation loops, bias detection, and continuous improvement frameworks.
arg-diagram
ARG academic-paper diagram mode — standalone structural & conceptual diagram generation.
graphite-morphology-classify
Classify graphite in a cast-iron micrograph per ASTM A247 / ISO 945-1, quantify nodularity, and read the matrix — the single most diagnostic observation in a cast-iron case.