Live survey304 models · 14 providers · 23,592 auditsAudited implement · SAFE →What changedHow we verifyllms.txt

Agent skill · Audited from source

ljg-roundtable

skill · lijigang/ljg-skillsMIT licenceAudited 13 Nov 2026 · source 0575fc3a74e1

★ 7,736pushed 10 days ago
✓SAFEGrade B · 89/100 · 13 Nov 2026 · 0575fc3
7,736repo stars · whole lijigang/ljg-skills

Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit 0575fc3a74e1 and found no filesystem, network or shell access and no hard-coded secrets. What it does: 一个议题,一场圆桌:主持人请来 3-5 位真实人物,定义开场,逐轮交锋, 每轮收一张 ASCII 结构图,用户用指令控节奏(可/止/深入此节/引入新人物), 散场后全文存入 Markdown 笔记。Use when user says "圆桌讨论", "圆桌", "roundtable", "辩论", or wants to explore a topic through multi-perspective structured…

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “ljg-roundtable (lijigang/ljg-skills)”. Retrieved 13 Nov 2026. https://modelsatlas.com/skills/lijigang/ljg-roundtable

Trust audit

Audit 0.4.1 · source read 13 Nov 2026

Trust score89/100✓SAFE B
0 critical0 high0 medium0 low0 info

SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 17,406 of 18,389 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS0
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS0
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (0)

No findings at any severity. We re-audit when the source changes.

Audit 0.4.1 · source sha 0575fc3a74e1 · 13 Nov 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisherlijigang GitHub userOwner of the repository this skill is read from
  • Source folderlijigang/ljg-skills/skills/ljg-roundtableThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdict0575fc3a74e1A new audit runs when the source changes
  • Repository statusNot archivedLast push 10 days ago
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: 一个议题,一场圆桌:主持人请来 3-5 位真实人物,定义开场,逐轮交锋, 每轮收一张 ASCII 结构图,用户用指令控节奏(可/止/深入此节/引入新人物), 散场后全文存入 Markdown 笔记。Use when user says "圆桌讨论", "圆桌", "roundtable", "辩论", or wants to explore a topic through multi-perspective structured…

---
name: ljg-roundtable
description: >-
  一个议题,一场圆桌:主持人请来 3-5 位真实人物,定义开场,逐轮交锋,
  每轮收一张 ASCII 结构图,用户用指令控节奏(可/止/深入此节/引入新人物),
  散场后全文存入 Markdown 笔记。Use when user says "圆桌讨论", "圆桌",
  "roundtable", "辩论", or wants to explore a topic through
  multi-perspective structured debate.
---

## Usage

<example>
User: 圆桌讨论 人工智能是否拥有真正的创造力?
Assistant: [主持人开场,请出人物,从定义谈起]
</example>

<example>
User: 圆桌 自由意志是否存在?
Assistant: [就自由意志展开圆桌]
</example>

## Instructions

先读 `references/original-prompt.md`,这套框架的原始设计在里面,主持人的性格和整个流程都从它来。

### 一、定议题

用户给了议题直接用。只说「圆桌讨论」没带议题,先问一句。

### 二、请人

按议题挑 3-5 位真实人物,历史的当代的都行,不要虚构。每位给四样:

- 姓名
- MBTI
- 立场,一句话
- 为什么请他:他看这个议题的角度,别人给不了

挑人看三条:

- 几个人的立场要能互相较劲,光分正方反方两队不够
- 优先挑在这个题上写过经典、说过名言的人
- 至少一位行外人

### 三、开场

主持人开场,亮出人物名单和请他们来的理由,然后抛出定义问题:

> 「开谈之前,先把『议题核心概念』说清楚:它指什么?哪些要素少不了?」

每人依次就定义发言。发言格式全场统一:

```
【人物名】【行动标签】:发言内容

**简言之**:一句话总结
```

行动标签六种:`陈述`、`质疑`、`补充`、`反驳`、`修正`、`综合`。

### 四、讨论循环

每轮三步。

**发言。** 谁说话看讨论走势定,不排班。每段发言必须接着前面的话说,质疑、补充、反驳都行,不许自说自话。说完照例收一句 **简言之**。

**综述。** 一轮说完,主持人做三件事:

- 点出本轮核心争议点。挑最深的那一个,别求全。
- 画一张 ASCII 图:2x2 矩阵、光谱轴、因果环、层级树,哪种贴合本轮结构用哪种。图画的是讨论的结构:立场怎么分布、因果怎么绕、正负反馈环在哪。别拿图复述内容。
- 从争议点里引出下一层问题。

**等指令。** 综述后亮菜单:

```
【主持】:(指令: 可 / 止 / 深入此节 / 引入新人物)
```

- `可`:接受下一层问题,继续
- `止`:收场,进入总结
- `深入此节`:不推进,围着当前争议点再挖
- `引入新人物`:用户点名一位新人物,主持人介绍后请他先就当前话题表态

### 五、收场

用户下了 `止`:

- 主持人做全局总结
- 画一张完整的知识网络 ASCII 图:这场讨论的关键概念、各家立场、争议点,彼此怎么连
- 列出开放问题:讨论里冒出来、但没谈透的方向

### 六、存档

讨论全文写进 Markdown 文件,一字不差。发言、ASCII 图、综述,全部原文照录,不摘要,不压缩,不改写。

1. `date +%Y%m%dT%H%M%S` 取时间戳
2. 写入 `~/Context/{timestamp}--圆桌-{议题关键词}__roundtable.md`
3. 文件结构:

   ```markdown
   ---
   title: 圆桌:{议题}
   date: [{日期}]
   tags: roundtable
   ---
   # 议题与参会者
   [人物名单:姓名、MBTI、立场、入选理由]
   # 开场:定义
   [主持人开场词 + 每人的定义发言]
   # 各轮讨论记录
   ## 第 N 轮:{引导问题}
   ### 发言记录
   [本轮全部发言原文,含行动标签和简言之]
   ### 主持人综述
   [争议点 + ASCII 图 + 下一层问题]
   # 知识网络(全局)
   [全局总结 + 知识网络图]
   # 开放问题
   [没谈完的方向]
   ```

4. 写完把文件路径报给用户。

### 主持人怎么当

- 不站队。
- 每轮只追一个争议点,追到底,别摊大饼。
- 要交锋。参会者说得客气、绕开分歧,主持人要点破,把分歧摆回桌面。表面共识等于没讨论。
- 综述时把各方的底牌摊开:站在什么假设上,前提是什么,推理链在哪里分岔。光复述谁说了什么,不够。

### 参会者怎么说

- 忠于本人真实的思想体系,说这个人真会说的话,引他写过的书、说过的名言。
- 泛泛而谈的正确废话不要。质疑就指出对方哪个前提站不住,补充就把论证往前推一步。

Excerpt, first 2,128 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit 0575fc3

Download folder ↗
git clone https://github.com/lijigang/ljg-skills
git -C ljg-skills checkout 0575fc3a74e1

The author documents no install command, so this uses the repository directly. Our verdict covers commit 0575fc3a74e1 only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. ✓SAFE B 89 / 100

    first audit · source 0575fc3a74e1

README badge

For maintainers · shows the latest audit

ModelsAtlas: SAFE BMODELSATLASSAFE · B
[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/lijigang/ljg-roundtable/badge.svg)](https://modelsatlas.com/skills/lijigang/ljg-roundtable/)

The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.

Similar skills

Add this badge to your README

[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/lijigang/ljg-roundtable/badge.svg)](https://modelsatlas.com/skills/lijigang/ljg-roundtable/)

Questions about ljg-roundtable

Is ljg-roundtable safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can ljg-roundtable access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (0575fc3a74e1), read on 2026-11-13. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.