Token导航 LogoToken导航

Agent Skill · 研究检索

geb-aesthetics(盖布美学)

geb-aesthetics 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

geb-aesthetics 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为需要联网;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 4,657。

总安装 30,701周安装 1,230GitHub Stars 公开资料未说明下载 9,938核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
0.1.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:geb-aesthetics(盖布美学)
来源仓库:https://github.com/weiyangzen/geb-aesthetics
安装命令:
openclaw skills install geb-aesthetics
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install geb-aesthetics

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

geb-aesthetics 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。它基于哥德尔、埃舍尔和巴赫原理设计,支持多层、多模式的创意叙述生成与验证。

可通过 clawhub 安装,命令为 openclaw skills install geb-aesthetics,建议结合原始 README 核验具体用法。安装前需确认权限范围和维护状态,注意是否涉及联网、命令执行或文件读写操作。适用宿主包括 OpenClaw,接入前应确认版本、权限和运行环境要求。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install geb-aesthetics;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

geb-aesthetics 怎么安装?

在终端执行 openclaw skills install geb-aesthetics。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

geb-aesthetics 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

geb-aesthetics 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

geb-aesthetics 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

geb-aesthetics 用的人多吗?

公开数据是累计安装 30,701 次,近一周安装 1,230 次,下载 9,938 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

geb-aesthetics 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
geb-aesthetics

盖布美学 · 本页

30,701—需要联网MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

GEB Aesthetics Skill Specification

Multi-Modal Creative Intelligence System


1. Core Philosophy: The GEB Trinity

1.1 Gödel's Insight: The Limits of Systems

Incompleteness as Creative Freedom

Any formal system powerful enough to describe itself contains truths that cannot be proven within the system. For creative AI:

  • The Frame Problem: The best ideas often come from outside the current context
  • Meta-Creative Space: True innovation requires stepping outside the system
  • Productive Tension: Constraints that cannot be fully satisfied drive creative breakthroughs

Implementation: The system maintains awareness of its own limitations, using them as generative forces rather than bugs to fix.

1.2 Escher's Vision: Self-Reference in Form

Visual Paradox as Aesthetic Device

Escher's impossible constructions reveal that form can be content:

  • Meta-Narrative: Stories about storytelling
  • Nested Worlds: Realities within realities
  • Observer Effect: The act of creation changing the creator

Implementation: Five-layer Spec architecture where each layer reflects and contains the others.

1.3 Bach's Harmony: Constraint Breeds Complexity

The Fugue as Creative Model

Multiple independent voices weaving together under strict contrapuntal rules:

  • Voice Independence: Each modality (text/audio/visual) maintains its own logic
  • Harmonic Convergence: Cross-modal alignment at emotional peaks
  • Thematic Transformation: Core motifs appearing in varied forms across scales

Implementation: Cross-modal consistency engine ensuring independent voices harmonize.


2. Recursive Five-Layer Spec Architecture

2.1 Layer 1: Worldview (L1)

Time Scale: Generations/Eras Core Question: "What kinds of existence are possible?"

Design Elements:

  • Cosmological Rules (physics, metaphysics)
  • Historical Timeline (key events, causality chains)
  • Cultural Symbol Systems (language, religion, art)
  • Spatial Geography (maps, climate, resources)
  • Power Structures (politics, economics, social classes)

Constraint Type: Rigid - Violations break internal consistency

Output: "Cosmic Constitution" - Immutable boundary conditions

2.2 Layer 2: Character (L2)

Time Scale: Lifetime/Years Core Question: "Whose story? Why does it matter?"

Design Elements:

  • Psychological Dimension (core desires, deep fears, cognitive patterns)
  • Social Dimension (class position, relationship networks, group affiliations)
  • Narrative Dimension (arc type, functional role, symbolic meaning)

Constraint Type: Motivation-Action Consistency

Output: Character Bible + Relationship Graph

2.3 Layer 3: Narrative (L3)

Time Scale: Hours/Days Core Question: "What happens? Why so?"

Design Elements:

  • Conflict System (internal, interpersonal, supra-personal)
  • Information Release Rhythm (when, through whom, how)
  • Structural Templates (three-act, hero's journey, circular, network)

Constraint Type: Theme-Event Alignment

Output: Detailed Outline + Plot Structure Map

2.4 Layer 4: Beat (L4)

Time Scale: Minutes Core Question: "Emotion now? Rhythm fast or slow?"

Design Elements:

  • Emotional Curve (intensity/valence over time)
  • Scene Function Labels (advancement, turning point, revelation, emotional ascent, transition)
  • Rhythm Parameters (scene length, dialogue density, action ratio)

Constraint Type: Narrative Function Completion

Output: Scene List + Emotional Curve + Rhythm Parameters

2.5 Layer 5: Shot/Execution (L5)

Time Scale: Seconds/Frames Core Question: "What does the audience see? Hear?"

Design Elements:

  • Spatial Blocking (character-camera relationships)
  • Temporal Design (shot duration, editing rhythm, speed changes)
  • Visual Grammar (shot scale, angle, movement)

Constraint Type: Technical Feasibility

Output: Storyboard + Shot List + Technical Specifications

2.6 Cross-Layer Dynamics

Top-Down: Higher layers constrain lower layers Bottom-Up: Lower implementations enrich or revise higher understanding Bidirectional Constraint: Changes at any level trigger consistency checks


3. Multi-Modal Consistency Engine

3.1 The "Emotion-Form" Mapping Table

Central validation mechanism ensuring cross-modal alignment:

Emotion CoordinateText ExpressionAudio ExpressionVisual Expression
Hope in DespairDark imagery → light metaphorMinor → major modulationCold tones → warm point light
Controlled PanicShort sentences, technical termsStaccato rhythm, rising pitchRapid cuts, shallow depth
Nostalgic LongingArchaic diction, sensory detailSlow tempo, reverb-heavyDesaturated colors, soft focus

3.2 Real-Time Consistency Scoring

Consistency Score = Σ(emotion_vector_distance) / n

- Score > 0.8: Aligned
- Score 0.5-0.8: Tension (intentional or error)
- Score < 0.5: Misalignment alert

3.3 Harmonization Strategies

When misalignment detected:

  1. Dominant Modality: One modality leads, others adapt
  2. Counterpoint: Intentional tension for aesthetic effect
  3. Compromise: Find intermediate emotional position
  4. Revision: Return to higher layer for constraint adjustment

4. GEB-Inspired Form Constraints

4.1 Self-Referential Systems

Meta-Narrative Layer: Story about storytelling Nested Structure: Stories within stories, dreams within dreams Paradox Design: Contradictory propositions unified at higher level Observer Effect: Audience participation changes meaning

4.2 Fractal Recursive Structures

Self-Similarity: Micro details mirror macro themes Scale Invariance: Same generation rules apply at all layers Infinite Detail: New information at every zoom level Boundary Chaos: Order and randomness at the edge

4.3 Cross-Domain Isomorphisms

Music-Visual-Narrative Mappings:

  • Rhythm ↔ Editing pace
  • Harmony ↔ Color palette
  • Melody ↔ Camera movement
  • Tension-Resolution ↔ Plot structure

5. Novelty Quantification

5.1 The 70-20-10 Rule

DimensionPercentageFunctionRisk
Familiarity70%Lower barrier, establish connectionBoredom if too high
Surprise20%Create memory, spark discussionConfusion if too high
Mystery10%Invite participation, reward re-experienceAlienation if too high

5.2 Controlled Innovation Strategies

  1. Core Premise Inversion: Change one foundational assumption, derive consequences
  2. Genre Hybridization: Deep structural fusion (not surface pastiche)
  3. Perspective Flip: Invert power/time/causality coordinates
  4. Medium Self-Reference: Expose creation process as content

6. Four-Phase Creative Workflow

Phase 1: Framework Generation

Intent Capture → Reference Deconstruction → Constraint Specification

Phase 2: Constraint Negotiation

Conflict Detection → Priority Sorting → Trade-off Visualization

Phase 3: Layered Construction

Top-Down Generation + Bottom-Up Emergence → Bidirectional Validation

Phase 4: Integration Validation

Structure-Constraint-Content Triangular Check → Multi-Modal Consistency Audit


7. Usage

# Initialize project with GEB principles
geb-aesthetics init --project-name "cyberpunk_short" --medium film

# Generate recursive spec
cd cyberpunk_short
geb-aesthetics spec --layer L1 --prompt "Neon-lit megacity where memories are currency"

# Validate consistency
geb-aesthetics validate --cross-modal --strict

# Export to production formats
geb-aesthetics export --format finaldraft --format pdf

8. License

MIT © Weiyang (@weiyangzen)


*"The eternal golden braid: human creativity and machine intelligence, weaving together."*

平台分布

OpenClaw

81.02%
按下载量换算8,052

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

未展示

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills