Ljg Paper Flow

Paper deep-reads start in Idea research when exploring ML, product, or strategy literature before validation or build commitments. The canonical shelf is research because the workflow’s first value is structured paper comprehension, not shipping product code.

Also useful

Also useful

Where it fits

Example use

Batch-read five arXiv agent papers into org notes and sketchnotes before picking a product direction.

Example use

ValidateScope & plan

Summarize competitor academic work into cards while scoping an MVP feature set.

Example use

Turn foundational ML papers into org files and PNGs for internal architecture references.

Example use

Produce visual paper cards for a newsletter or social thread on recent research.

How it compares

Orchestration workflow across two LJG skills—not a standalone PDF parser or a generic literature-review MCP.

Common Questions / FAQ

Who is ljg-paper-flow for?

Solo builders and researchers using the LJG skill stack who want multi-paper reading plus card generation from one user-invoked workflow.

When should I use ljg-paper-flow?

Use it during Idea research when exploring papers, during Build docs when capturing design references, or during Grow content when turning papers into visual assets—trigger on 论文流, paper flow, or multiple URLs needing analysis and cards.

Is ljg-paper-flow safe to install?

It orchestrates other skills and may fetch external papers; check the Security Audits panel on this Prism page and review child skill permissions before enabling network-heavy runs.

Workflow Chain

Requires first: ljg paper, ljg card

SKILL.md

READMESKILL.md - Ljg Paper Flow

# ljg-paper-flow: 论文流

一条命令完成：读论文 → 生成解读 → 铸成卡片。支持多篇并行。

## 模式

**强制 NATIVE 模式。** 本 workflow 是纯 skill 管道（ljg-paper → ljg-card），不需要 Algorithm 的七步流程。直接按下方执行步骤调用 skill，不走 OBSERVE/THINK/PLAN/BUILD/EXECUTE/VERIFY/LEARN。

## 参数

| 参数 | 说明 |
|------|------|
| 无参数 | 对话中已提供的论文链接/文件 |
| `-l` | 卡片模具改用长图模式（默认 `-v` 视觉笔记） |
| `-i` | 卡片模具改用信息图模式 |
| `-c` | 卡片模具改用漫画模式 |

## 执行

### 1. 收集论文列表

从用户消息中提取所有论文来源（arxiv URL、PDF 路径、论文名称等）。

### 2. 并行处理每篇论文

对每篇论文，启动一个 Agent subagent，每个 subagent 按顺序执行两步：

**步骤 A — 读论文（ljg-paper）：**

调用 Skill tool 执行 `ljg-paper`，传入该论文的来源。等待完成，获得生成的 org 文件路径。

**步骤 B — 铸卡片（ljg-card）：**

读取步骤 A 生成的 org 文件，调用 Skill tool 执行 `ljg-card`（默认 `-v`，或按用户指定的模具参数），以 org 文件内容为输入。等待完成，获得 PNG 文件路径。

### 3. 汇总报告

所有论文处理完成后，汇总输出：

```
════ 论文流完成 ═══════════════════════
📄 {论文标题1}
   📝 解读: {org 文件路径}
   🖼️ 卡片: {PNG 文件路径}

📄 {论文标题2}
   📝 解读: {org 文件路径}
   🖼️ 卡片: {PNG 文件路径}
...
```

## 关键约束

- 每篇论文的两步必须串行（先 paper 后 card），但多篇论文之间并行
- ljg-paper 和 ljg-card 各自的质量标准、红线、品味准则不变
- 卡片内容来自生成的 org 文件，不是原始论文

What is this skill?

Forced NATIVE skill pipeline—skips Algorithm OBSERVE/THINK/PLAN seven-step ritual

Per paper: serial ljg-paper then ljg-card; multiple papers processed in parallel subagents

Card mold flags: default -v visual sketchnote, optional -l long image, -i infographic, -c comic

Aggregated 论文流 completion report with org paths and PNG paths per title

2-step serial pipeline per paper (ljg-paper → ljg-card)

4 card mold modes including default -v visual sketchnote

Compatible agents: Claude Code, Cursor, Codex, any compatible agent

Adoption & trust: 3.4k installs on skills.sh; 5.5k GitHub stars; 2/3 security scanners passed (skills.sh audits).

Journey fit

Spans multiple journey phases - primary shelf plus alternate fits below.

Primary fit

Also useful

Also useful

Where it fits

Example use

Batch-read five arXiv agent papers into org notes and sketchnotes before picking a product direction.

Example use

ValidateScope & plan

Summarize competitor academic work into cards while scoping an MVP feature set.

Example use

Turn foundational ML papers into org files and PNGs for internal architecture references.

Example use