GitHub - chansigit/studypaper: paper-deepstudy: deep paper study plugin for ML and computational biology

Read it. Roast it. Reproduce it.

One PDF in. A peer-review-grade workspace out — in twenty minutes.

Install · 安装

In any Claude Code session (CLI / IDE / Web), run:

/plugin marketplace add chansigit/studypaper
/plugin install paperstudio@studypaper

Requires claude-paper installed first — the marketplace does not auto-install dependencies yet. · 需先装 claude-paper。

English · 中文 · Design · Quick start · Examples · Changelog

English

You have 50 papers in your reading queue. You scan abstracts. You skim figures. You forget what you read by Friday.

studypaper does the work you wish you had time for.

It is a Claude Code plugin that turns any ML or computational-biology paper (PDF or arXiv URL) into a complete, navigable research workspace — a structured analysis directory, a reviewer-style verdict, and bilingual social-media notes — from one command, in about twenty minutes. Run extension commands (adversarial review, deep-dive, head-to-head compare, reproducibility audit) on top.

Why studypaper

Without studypaper	With studypaper
You scan abstracts and forget what you read	A 7-file structured analysis you can `grep`, diff, and revisit a year later
You vaguely sense "this paper is sketchy"	An adversarial review captures the objection in writing — author defense, blind judge, your call
You assume the GitHub link works	A 7-dimension reproducibility audit verifies it live and flags missing seeds, hardware, eval scripts
You write the WeChat / 小红书 post from scratch	A 3000-character draft lands ready, with auto-selected figures and 5 candidate titles
Your second paper repeats the work of your first	Re-usable domain packs (`single-cell`, `genomics`, `protein-structure` …) inject the right context every time

What you get

Every /paperstudio:study produces these artifacts under ~/claude-papers/papers/<slug>/:

analysis/
  00-paper-profile.md       paper type · domain · difficulty (YAML frontmatter)
  01-problem.md             problem statement and framing
  02-formalization.md       math: notation, loss, constraints
  03-method-deep.md         method with rationale + alternatives considered
  04-experiments.md         experiment critique (not just description)
  05-prior-work.md          chronological timeline + comparison
  06-figures.md             per-figure interpretation + scoring
review.md                   academic-reviewer-style verdict (Strengths / Weaknesses / Score)
notes/
  source.md                 unified Chinese source (single point of truth)
  titles.md                 5+5 candidate titles
  xhs.md                    Xiaohongshu rendering (~1000 chars, 1 figure)
  wechat.md                 WeChat rendering (~3000 chars, 2-3 figures)

The remaining workspace artifacts are produced by extension commands, not by /paperstudio:study:

Command	Artifact
`/paperstudio:review-round`	`review-rounds/round-NN-<title>.md` (one file per round)
`/paperstudio:deep-dive`	`deep-dives/<topic-slug>.md`
`/paperstudio:compare`	`compares/vs-<other-slug>.md`
`/paperstudio:reproduce-check`	`reproduce-check.md`

Every file is regeneratable. Every mutation backs up to <file>.bak.NN. Nothing is destructive.

Install

In Claude Code (CLI / IDE / Web):

/plugin marketplace add chansigit/studypaper
/plugin install paperstudio@studypaper

Prerequisites

claude-paper plugin installed (declared as a dependency; install it first — the marketplace does not auto-install dependencies yet).
pdftotext (from poppler-utils) on PATH for full-text extraction. Optional — without it, the orchestrator falls back to passing the PDF directly to sub-agents.
- macOS: brew install poppler
- Debian/Ubuntu: sudo apt install poppler-utils

Design · prompt-as-code

paperstudio is not a typical software project — it is an LLM orchestration system written almost entirely in Markdown. The "code" is constraints, templates, and dispatch instructions that the model interprets at runtime. Understanding it requires switching frames: Markdown is the program; the model is the runtime.

Five layers, top to bottom:

[1] commands/      ×10   Slash-command entry points. Just argument hints +
                         "go run that skill". No business logic.
[2] skills/        ×9    The orchestrators ("directors"). Stage ordering, flag
                         parsing, idempotence rules, who-dispatches-whom.
                         This is where the project's control flow lives.
[3] prompts/       ×19   Sub-agent scripts ("actors"). Each one is the full
                         prompt for a single specialized sub-agent (paper-profiler,
                         method-analyst, reviewer-synthesizer, …). They never
                         talk to each other directly — they exchange data via
                         files on disk that the skill layer routes.
[4] templates/     ×16   Output skeletons with <runtime-timestamp>,
                         <fill-this> placeholders. Sub-agents copy + fill.
                         Job: enforce uniform structure (required H2 sections,
                         frontmatter keys) so downstream parsing is reliable.
[5] scripts/             Deterministic bits (cjs + sh): things the model
                         shouldn't do (URL normalization, schema validation,
                         JSONL log writing, figure ranking, slugification).

Two filesystem coordinate systems — keep them straight:

	Plugin code (read-only, in Git)	User data (read/write, not in Git)
Location	`~/.claude/plugins/cache/studypaper/paperstudio/<ver>/`	`$CLAUDE_PAPERS_ROOT` (default `~/claude-papers/papers/<slug>/`)
Owns	the prompts, templates, scripts	the paper, all generated artifacts, dispatch log
Stable	yes (versioned)	no (one folder per paper studied)

Four cross-cutting abstractions you'll see everywhere:

Provenance. Every generated file's line 1 is . Audit-by-grep: any artifact tells you who, when, with which version. Enforced by validate-artifact.sh.
Idempotence. If an output already exists, the dispatching skill skips that sub-agent (no token re-burn). --force backs up to <file>.bak.NN before overwrite. --only <stage> re-runs just one stage. The whole pipeline is safe to interrupt and resume.
Dispatch log. <paper>/.deepstudy/run.jsonl — one JSONL line per sub-agent dispatch, with status + timestamp + plugin version. The only runtime observability signal. Local only; opt out with PAPER_DEEPSTUDY_NO_RUN_LOG=1.
Domain packs. domain-packs/{single-cell,protein-structure,ml-pure,…}.md are not docs — they are knowledge-injection text. The paper-profiler picks 1–N relevant packs at Stage 0; later sub-agents prepend the chosen pack contents to their own prompts. Multi-domain adaptation by concatenation, not by maintaining per-domain prompt forks.

Walking through one /paperstudio:study call:

user types  /paperstudio:study https://arxiv.org/abs/1706.03762
                  │
                  ▼
[command]  commands/study.md  ──── dispatches ───▶  Skill(study-deep)
                  │
                  ▼
[skill]    skills/study-deep/SKILL.md  (the orchestrator wakes up)
            ├── 0.1  verify-prereqs.sh
            ├── 0.2  normalize URL → claude-paper:study downloads PDF
            ├── 0.3  resolve paper folder, set $PAPER_DIR
            ├── 0.4  Agent(paper-profiler)  → analysis/00-paper-profile.md
            │
            ├── Stage 1 (6 parallel Agent calls, one prompt each):
            │     problem-framer / formalizer / method-analyst /
            │     experiment-critic / prior-work-historian / figure-interpreter
            │     → analysis/01–06
            │
            ├── Stage 2: Agent(reviewer-synthesizer) → review.md
            │
            └── Stage 3:
                  Agent(notes-writer)     → notes/source.md
                  Agent(title-generator)  → notes/titles.md
                  select-figures.cjs      (deterministic figure pick)
                  Agent(xhs-renderer) ║ Agent(wechat-renderer)  parallel
                  → notes/xhs.md, notes/wechat.md

After every Agent call: log_dispatch <agent> <output> ok appends a JSONL line. After every Stage: idempotence rule decides skip-or-overwrite.

Why this architecture is worth it:

Iterating on model behavior = editing Markdown, not redeploying code.
Source and docs are the same file — onboarding is reading the prompts.
Layers have crisp responsibilities → small bug-fix radius.

What you give up:

No stack traces. Debugging = reading .deepstudy/run.jsonl + diffing produced artifacts.
No static guarantees. Correctness = constraints in prompts + schema validation + golden-snapshot behavior tests (tests/behavior/). The discipline of catching drift falls on the test suite.
A prompt change can silently shift output style across the whole pipeline. The 17-assertion golden-snapshot test is the safety net for this.

Where to extend:

You want to add…	Touch this
A new slash command	`commands/<name>.md` + `skills/<name>/SKILL.md`
A new sub-agent role	`prompts/<role>.md` + dispatch it from some `SKILL.md`
A new domain	`domain-packs/<x>.md` + add detection rule to `paper-profiler.md`
A new artifact type	`templates/<x>.md` + new schema arm in `validate-artifact.sh`
A new paper-host URL	regex case in `scripts/normalize-paper-url.sh`

If you only remember one thing: constraints + templates + schema validation + behavior tests are how this project tames an otherwise unpredictable runtime (the model). Take any one of those four away and the whole thing collapses into "creative drift".

Quick start

# One-shot full pipeline — fetch, analyze, review, render notes
/paperstudio:study https://arxiv.org/abs/1706.03762

# Same, but render the Stage 3 notes (xhs / wechat) in English instead of the default 中文
/paperstudio:study https://arxiv.org/abs/1706.03762 --lang en

# An adversarial review round — you raise an objection, defense and blind judge respond
/paperstudio:review-round

# Drill into a sub-topic that the analysis brushed over
/paperstudio:deep-dive "scaled dot-product attention derivation"

# Head-to-head with another paper you've already studied (or auto-study + compare)
/paperstudio:compare attention-is-all-you-need --lang zh

# 7-dimension reproducibility audit, with live GitHub link verification
/paperstudio:reproduce-check

Commands at a glance

Command	What it does
`/paperstudio:study <pdf-or-url> [--lang en]`	One-shot full pipeline. Default Stage 3 (xhs / wechat) language is 中文; `--lang en` switches to English.
`/paperstudio:rerun-stage <stage>`	Re-run a single stage (`analysis` / `review` / `notes` / `profile`)
`/paperstudio:review-round`	Adversarial review round (objection → defense → blind judge → user verdict)
`/paperstudio:refine-notes <variant>`	Apply an edit instruction to `xhs.md` or `wechat.md`
`/paperstudio:retitle <variant>`	Regenerate 5 title candidates
`/paperstudio:reselect-figures`	Re-pick which figures get embedded
`/paperstudio:deep-dive <topic>`	Focused sub-topic write-up
`/paperstudio:compare <target>`	Head-to-head comparison with another paper
`/paperstudio:add-prior-work <ref>`	Append a missed prior-work entry (arXiv URL / BibTeX)
`/paperstudio:reproduce-check`	7-dimension reproducibility audit

Run any command without arguments for inline help, or see paperstudio/README.md for the full reference.

Examples

Real outputs from running the pipeline on actual papers:

examples/string-database-2025/ — full pipeline on The STRING database in 2025 (a cs-bio / protein-function database paper). Includes the adversarial review round, the deep-dive, the cross-paper comparison, the reproducibility audit, and the bilingual notes — every artifact generated by the live integration test.

Repository layout

studypaper/
├── .claude-plugin/
│   └── marketplace.json          marketplace registration — what makes /plugin install work
├── paperstudio/              the plugin
│   ├── .claude-plugin/plugin.json
│   ├── commands/                 10 slash commands
│   ├── skills/                   orchestration skills (study-deep, review-round, …)
│   ├── prompts/                  19 sub-agent prompts
│   ├── templates/                output templates for every artifact
│   ├── domain-packs/             7 domain knowledge packs
│   ├── scripts/                  helper scripts (verify-prereqs, parse-judge-output, …)
│   └── tests/                    225 bats + 4 node + integration smoke
├── examples/                     curated real-paper outputs
├── assets/                       logo + banner SVGs
└── docs/                         design specs and implementation plans

Contributing

The project follows test-driven development. Run the suite:

cd paperstudio
npm install      # one-time, installs Bats
npm run test:unit

Structural assertions are bats-based; pure-logic helpers have node test scripts. The integration smoke test (tests/integration/test-end-to-end.sh) verifies file-level wiring without dispatching real sub-agents.

For non-trivial changes, the project uses the Superpowers workflow: brainstorming → spec → plan → subagent-driven implementation. Specs live in docs/superpowers/specs/; plans in docs/superpowers/plans/.

License

MIT — see LICENSE.

Credits

Built on top of claude-paper by alaliqing. Workflow patterns (TDD, subagent-driven development, brainstorming) come from the superpowers skills library. Logo and banner crafted in plain SVG.

中文

你的待读论文堆了 50 篇。你扫摘要、瞄图、礼拜五就忘了自己看过啥。

studypaper 替你做你一直没时间做的事情。

它是一个 Claude Code 插件,把任意一篇机器学习或计算生物学论文(PDF 或 arXiv 链接)转换成一个完整、可导航的研究工作区 —— 一份结构化分析目录、一份审稿人视角的判定、一套双语社交媒体笔记 —— 一条命令搞定,大约二十分钟。扩展命令(对抗性审阅、深挖、正面对比、可复现性审计)按需追加。

为什么是 studypaper

没有 studypaper	有 studypaper
扫摘要,扫完就忘	一份 7 文件结构化分析,可 `grep`、可 diff、一年后还能回看
隐约觉得"这论文有点水"	把质疑写下来 —— 作者辩护、盲审 judge、你最终拍板
默认 GitHub 链接还活着	7 维可复现性审计,实时验证链接、标注缺失的种子/硬件/评估脚本
微信 / 小红书帖子从零写	一份 3000 字草稿现成,配图自动选好,5 个候选标题候选
第二篇论文重复第一篇的功夫	可复用的领域包(`single-cell`、`genomics`、`protein-structure` …)每次自动注入对应上下文

你会得到什么

每次 /paperstudio:study 在 ~/claude-papers/papers/<slug>/ 下生成以下产物:

analysis/
  00-paper-profile.md       论文类型 · 领域 · 难度(YAML frontmatter)
  01-problem.md             问题陈述与框定
  02-formalization.md       数学:符号、损失、约束
  03-method-deep.md         方法精读 + 设计 rationale + 候选方案
  04-experiments.md         实验批评(不仅是描述)
  05-prior-work.md          时间线 + 对比
  06-figures.md             逐图解读 + 评分
review.md                   学术审稿人风格判定(优点 / 缺点 / 分数)
notes/
  source.md                 中文统一 source(唯一真源)
  titles.md                 5+5 候选标题
  xhs.md                    小红书渲染(~1000 字,1 张图)
  wechat.md                 微信渲染(~3000 字,2-3 张图)

其余工作区产物由扩展命令生成,不属于 /paperstudio:study:

命令	产物
`/paperstudio:review-round`	`review-rounds/round-NN-<title>.md`(每轮一个文件)
`/paperstudio:deep-dive`	`deep-dives/<topic-slug>.md`
`/paperstudio:compare`	`compares/vs-<other-slug>.md`
`/paperstudio:reproduce-check`	`reproduce-check.md`

每个文件都可重新生成。任何修改前都备份成 <file>.bak.NN。无破坏性操作。

安装

在 Claude Code(CLI / IDE / Web)中执行:

/plugin marketplace add chansigit/studypaper
/plugin install paperstudio@studypaper

前置要求

已安装 claude-paper 插件(声明为依赖,但目前 marketplace 不会自动装依赖,需先手动安装)。
pdftotext(来自 poppler-utils)在 PATH 中用于全文抽取。可选 —— 缺失时 orchestrator 会退化为把 PDF 直接传给 sub-agent。
- macOS:brew install poppler
- Debian/Ubuntu:sudo apt install poppler-utils

设计 · prompt-as-code

paperstudio 不是常规软件项目 —— 它是一个几乎完全用 Markdown 写成的 LLM 编排系统。"代码"是约束、模板和派发指令,由模型在运行时解释。理解它需要切换思维:Markdown 是程序;模型是运行时。

自顶向下五层:

[1] commands/      ×10   斜杠命令入口。只有 argument-hint + "去跑下面那个 skill"。
                         不含业务逻辑。
[2] skills/        ×9    编排器("导演")。stage 顺序、flag 解析、幂等规则、
                         谁派给谁。项目控制流的所在。
[3] prompts/       ×19   子 Agent 脚本("演员")。每个文件是一个专职子 Agent
                         (paper-profiler / method-analyst / reviewer-synthesizer …)
                         的完整 prompt。它们彼此不直接通信 —— 通过磁盘文件
                         交接,skill 层负责路由。
[4] templates/     ×16   产物骨架,带 <runtime-timestamp> / <fill-this> 占位符。
                         子 Agent 复制并填空。目的:统一结构(必需的 H2 小节、
                         frontmatter key),让下游解析可靠。
[5] scripts/             确定性逻辑(cjs + sh):模型不擅长做的事
                         (URL 归一化、schema 校验、JSONL 日志、图表排序、
                         slug 生成)。

两套文件系统坐标 —— 必须分清:

	插件代码(只读,跟 Git)	用户数据(读写,不进 Git)
位置	`~/.claude/plugins/cache/studypaper/paperstudio/<ver>/`	`$CLAUDE_PAPERS_ROOT`(默认 `~/claude-papers/papers/<slug>/`)
拥有	prompts、templates、scripts	论文、所有产物、dispatch log
稳定性	是(版本化)	否(每篇论文一个文件夹)

4 个横切抽象,你会反复见到:

Provenance(留痕)。 每个产物 line 1 必为 。grep 即审计:任何文件都告诉你"谁、什么时候、用哪版插件"。validate-artifact.sh 强制。
幂等。 产物已存在 → skill 跳过对应子 Agent(不重烧 token)。--force 备份成 <file>.bak.NN 再覆盖。--only <stage> 只重跑一个 stage。整条流水线随时可中断、可续跑。
Dispatch log。 <paper>/.deepstudy/run.jsonl —— 每次子 Agent 派发追加一行 JSONL,带状态 + 时间戳 + 插件版本。唯一的运行时可观测信号。仅本地;PAPER_DEEPSTUDY_NO_RUN_LOG=1 可关。
Domain pack。 domain-packs/{single-cell,protein-structure,ml-pure,…}.md 不是文档,是知识注入文本。paper-profiler 在 Stage 0 选出 1–N 个相关 pack;后续每个子 Agent 把所选 pack 内容拼到自己 prompt 前面再发。多领域适配靠拼接,不靠维护多份 prompt 分支。

一次 /paperstudio:study 的调用走向:

用户敲     /paperstudio:study https://arxiv.org/abs/1706.03762
                  │
                  ▼
[命令]     commands/study.md  ──── 派发 ───▶  Skill(study-deep)
                  │
                  ▼
[skill]    skills/study-deep/SKILL.md  (orchestrator 醒来)
            ├── 0.1  verify-prereqs.sh
            ├── 0.2  归一化 URL → claude-paper:study 下载 PDF
            ├── 0.3  解析 paper folder,设置 $PAPER_DIR
            ├── 0.4  Agent(paper-profiler) → analysis/00-paper-profile.md
            │
            ├── Stage 1(6 个子 Agent 并行,每个一份 prompt):
            │     problem-framer / formalizer / method-analyst /
            │     experiment-critic / prior-work-historian / figure-interpreter
            │     → analysis/01–06
            │
            ├── Stage 2:Agent(reviewer-synthesizer) → review.md
            │
            └── Stage 3:
                  Agent(notes-writer)     → notes/source.md
                  Agent(title-generator)  → notes/titles.md
                  select-figures.cjs      (确定性选图)
                  Agent(xhs-renderer) ║ Agent(wechat-renderer)  并行
                  → notes/xhs.md, notes/wechat.md

每次 Agent 调用后:log_dispatch <agent> <output> ok 追加 JSONL。每个 Stage 后:幂等规则决定跳过还是覆盖。

为什么这种架构值得:

迭代模型行为 = 改 Markdown,不需重新部署代码
源码和文档是同一份文件,新人 onboard 就是读 prompt
各层职责清晰,bug 半径小

代价:

没有 stack trace。调试 = 读 .deepstudy/run.jsonl + diff 产物
没有静态正确性保证。正确性 = prompt 里的约束 + schema 校验 + golden-snapshot 行为测试(tests/behavior/)。抓漂移的纪律全押在测试套件上
prompt 一改,整条流水线产出风格可能静默漂移。17 条 golden-snapshot 测试是这条的安全网

扩展点:

想加什么	改哪
新的斜杠命令	`commands/<name>.md` + `skills/<name>/SKILL.md`
新的子 Agent 角色	`prompts/<role>.md` + 在某个 `SKILL.md` 里 dispatch 它
新领域	`domain-packs/<x>.md` + 在 `paper-profiler.md` 加识别规则
新的产物类型	`templates/<x>.md` + 在 `validate-artifact.sh` 加 schema arm
新的论文 host URL	在 `scripts/normalize-paper-url.sh` 加正则

只需记住一件事:约束 + 模板 + schema 校验 + 行为测试,这四件事一起驯服了一个本来不可预测的运行时(模型)。少了任何一个,整套就坍缩成"创意漂移"。

快速上手

# 一键全自动 —— 下载、分析、审稿、渲染笔记
/paperstudio:study https://arxiv.org/abs/1706.03762

# 同上,但 Stage 3 笔记(xhs / wechat)输出英文,而不是默认中文
/paperstudio:study https://arxiv.org/abs/1706.03762 --lang en

# 一轮对抗式审稿 —— 你提质疑,辩护方和盲审 judge 应答
/paperstudio:review-round

# 钻入一个分析没展开的子话题
/paperstudio:deep-dive "scaled dot-product attention 推导"

# 与另一篇已研读的论文做正面比较(或自动研读 + 比较)
/paperstudio:compare attention-is-all-you-need --lang zh

# 7 维可复现性审计,实时验证 GitHub 链接
/paperstudio:reproduce-check

命令一览

命令	用途
`/paperstudio:study <pdf-or-url> [--lang en]`	一键全自动 pipeline。Stage 3 笔记默认中文;`--lang en` 切换为英文。
`/paperstudio:rerun-stage <stage>`	重跑单个 stage(`analysis` / `review` / `notes` / `profile`)
`/paperstudio:review-round`	对抗式审稿(质疑 → 辩护 → 盲审 → 用户拍板)
`/paperstudio:refine-notes <variant>`	对 `xhs.md` 或 `wechat.md` 应用一条修改指令
`/paperstudio:retitle <variant>`	重新生成 5 个候选标题
`/paperstudio:reselect-figures`	重新选取嵌入哪些图
`/paperstudio:deep-dive <topic>`	子话题深度展开
`/paperstudio:compare <target>`	与另一篇论文正面对比
`/paperstudio:add-prior-work <ref>`	增补一条先前工作(arXiv URL / BibTeX)
`/paperstudio:reproduce-check`	7 维可复现性审计

不带参数运行任何命令可看 inline help,完整参考见 paperstudio/README.md。

示例

对真实论文跑完 pipeline 的实际产物:

examples/string-database-2025/ —— 在《The STRING database in 2025》(cs-bio / protein-function 数据库类论文)上的完整 pipeline。包含对抗审稿、深度展开、跨论文比较、可复现性审计、双语笔记 —— 全部由 live 集成测试生成。

仓库结构

studypaper/
├── .claude-plugin/
│   └── marketplace.json          marketplace 注册 —— 让 /plugin install 能识别的关键
├── paperstudio/              插件本体
│   ├── .claude-plugin/plugin.json
│   ├── commands/                 10 个 slash 命令
│   ├── skills/                   orchestration 技能(study-deep, review-round, …)
│   ├── prompts/                  19 个 sub-agent 提示词
│   ├── templates/                所有产物的模板
│   ├── domain-packs/             7 个领域知识包
│   ├── scripts/                  辅助脚本
│   └── tests/                    225 bats + 4 node + 集成 smoke
├── examples/                     精选真实论文产物
├── assets/                       logo + banner SVG
└── docs/                         设计 spec 和实现 plan

贡献

项目遵循 TDD。运行测试套件:

cd paperstudio
npm install      # 一次性,安装 Bats
npm run test:unit

结构性断言基于 bats;纯逻辑 helper 有 node 测试脚本。集成 smoke test 验证文件级 wiring,不会真的派 sub-agent。

非平凡改动遵循 Superpowers 工作流:brainstorming → spec → plan → subagent-driven 实现。Spec 在 docs/superpowers/specs/,plan 在 docs/superpowers/plans/。

许可

MIT —— 详见 LICENSE。

致谢

构建在 alaliqing 的 claude-paper 之上。工作流模式(TDD、subagent-driven 开发、brainstorming)来自 superpowers 技能库。Logo 与 banner 由纯 SVG 手写。

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

Read it. Roast it. Reproduce it.

Install · 安装

English

Why studypaper

What you get

Install

Design · prompt-as-code

Quick start

Commands at a glance

Examples

Repository layout

Contributing

License

Credits

中文

为什么是 studypaper

你会得到什么

安装

设计 · prompt-as-code

快速上手

命令一览

示例

仓库结构

贡献

许可

致谢

About

Uh oh!

Releases

Packages

Uh oh!

Contributors

Uh oh!

Languages

Name		Name	Last commit message	Last commit date
Latest commit History 161 Commits
.claude-plugin		.claude-plugin
assets		assets
docs/superpowers		docs/superpowers
examples		examples
paperstudio		paperstudio
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
LICENSE		LICENSE
README.md		README.md

Folders and files

Latest commit

History

Repository files navigation

Read it. Roast it. Reproduce it.

Install · 安装

English

Why studypaper

What you get

Install

Design · prompt-as-code

Quick start

Commands at a glance

Examples

Repository layout

Contributing

License

Credits

中文

为什么是 studypaper

你会得到什么

安装

设计 · prompt-as-code

快速上手

命令一览

示例

仓库结构

贡献

许可

致谢

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages