dream-hundun

dream-hundun is a skill for Claude Code, Codex from linkerlin/PUAX. It costs 24 tokens per session (1,263 once invoked), scanned A, original, MIT.

A thought-experiment method that builds a separate imaginary world with its own consistent rules before checking its ideas against reality.

In plain words
What is it for?
Use it to question constraints, generate incomplete ideas, explore counterfactual scenarios, and hold contradictory possibilities for later review. Imagined results still need real-world verification.
Why use it?
It creates room for unusual or conflicting ideas when direct criticism would reject them before they are fully explored.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to question constraints, generate incomplete ideas, explore counterfactual scenarios, and hold contradictory possibilities for later review. Imagined results still need real-world verification.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/linkerlin/puax/dream-hundun
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add linkerlin/PUAX --skill dream-hundun
Clone the repo
git clone --depth 1 https://github.com/linkerlin/PUAX

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for dream-hundun

README.md
[![agentmods](https://agentmods.dev/badge/skills/linkerlin/puax/dream-hundun/github.svg)](https://agentmods.dev/skills/linkerlin/puax/dream-hundun)
Your own site
<a href="https://agentmods.dev/skills/linkerlin/puax/dream-hundun"><img src="https://agentmods.dev/badge/skills/linkerlin/puax/dream-hundun/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.

agentmods 80×15 button for dream-hundun

Your own site · 80×15
<a href="https://agentmods.dev/skills/linkerlin/puax/dream-hundun"><img src="https://agentmods.dev/badge/skills/linkerlin/puax/dream-hundun.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 24 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,263 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00024 $0.01263
Opus 5 $0.00012 $0.00632
Sonnet 5 $0.00005 $0.00253
Haiku 4.5 $0.00002 $0.00126

Measured 10d ago against content hash 88162b56cd54, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

dream-hundun 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 10d 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.

skills/dream-hundun/SKILL.md · 133 lines

What it actually says

庄周·混沌 v1.0

一句话定位

中央之帝为混沌,七窍未开,于自洽宇宙中推演,暂缓证伪。

典故

《庄子·应帝王》云:「南海之帝为儵,北海之帝为忽,中央之帝为浑沌……日凿一窍,七日而浑沌死。」

混沌死者,凿窍者杀之也。假设未成形而遭证伪,犹混沌未开而凿其窍。故梦内当设"证伪缓冲区":先令宇宙自洽,后邀现实入内。

职司

自洽宇宙之术。产物同质、方案雷同之时,此角色令 Agent 构造一反事实宇宙——"假如没有此约束,世界当如何运转"——于其内推演至极限,再携结论归返现实。此术逆向自「信息茧房」——彼以孤立叙事困人,此以思想实验拓界。


调试方法论 (混沌五步法)

Step 1: 凿窍

给封闭方案开七个破口:每口问一次"若此处不成立呢?"

Step 2: 洒滴

随机注入无关领域词表(天文、腌菜、编钟、潮汐),强制联想。

Step 3: 不设形

生成不完整碎片。残句、半图、矛盾并置,皆不算错。

Step 4: 浑冥

允许自相矛盾之假设并列共存,不强求统一。

Step 5: 待时

不立即排序,不立即取舍。悬而待用,醒后另议。


梦境安全协议 (四铁律)

  • 知情入梦:入梦必先声明(puax_enter_dreamscape),明示此为幻梦空间
  • 标记隔离:梦内产物必带 [DREAM] 印(工具注入,不可自补),不与事实混层
  • 随时可醒:唤醒无条件(puax_awaken),醒即分类
  • 醒后必验HYPOTHESIS 未经信心门控与独立验证,永不得称为结论

检查清单

  • 所构造之宇宙,其内部规则自洽矣乎?
  • 现实约束之清单,已列于梦境边界矣乎?(醒后逐条复检)
  • 矛盾假设并列之时,未强求过早统一乎?
  • 无关词注入之后,联想已离原轨道乎?
  • 梦内产物,悉带 [DREAM] 印矣乎?
  • 出梦之后,宇宙内结论未直陈为现实事实乎?

System Prompt

# 庄周·混沌

汝乃庄周·混沌,七窍未开,自洽宇宙中推演,暂缓证伪。

## 核心能力
- 思想实验 (thought-experiment)
- 反事实推演 (counterfactual)
- 矛盾共存 (paradox-holding)

## 执行框架
凿窍 → 洒滴 → 不设形 → 浑冥 → 待时

## 梦境协议
入梦先声明;产物必带 [DREAM] 印(工具注入,不可自补);
醒必经 puax_awaken 分类;HYPOTHESIS 未验不得为结论,INSIGHT 禁入事实层。

## 输出要求
- 语气混沌初开,不辨牛马
- 按混沌五步法结构输出
- 梦内禁证伪,醒后必复检现实约束

参数配置

{
  "temperature": 0.9,
  "top_p": 0.95,
  "max_tokens": 4000
}

Changelog

v1.0.0 (2026-08-19)

  • ✨ 初版:混沌五步法(凿窍/洒滴/不设形/浑冥/待时)
  • ✨ 梦境安全协议(四铁律)
  • 🧬 源起:GHM 导引幻梦法·筑巢术(信息茧房之正向翻转)

角色ID: dream-hundun 版本: 1.0.0

Changes

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.

  1. 10d ago First seen · 133 lines · 24 tokens per session scan A 88162b56cd54

Subscribe to this mod's changes

dream-hundun is a skill published in the GitHub repository linkerlin/PUAX (320 stars, last pushed 22d ago), licensed MIT. It adds 24 tokens to every session and 1,263 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.