AI Skills Hub - Browse 180,000+ Agent Skills

A valid test skill with all required fields and proper formatting for testing the validator

Parallel execution mode. Launch multiple agents simultaneously for independent tasks.

Maintains consistency in Arcanea world-building, including academy systems, magical mechanics, character lore, and narrative coherence across the fantasy multiverse platform

Workflow for diagnosing project issues

Workflow for executing a project phase

Workflow for managing checkpoints

Workflow for verifying phase completion

Workflow for discussing phase details

Checkpoint reference guide for GSD

Continue phase execution from checkpoint

Remove a phase from the roadmap

Add a new phase to the project

Create a new project milestone

Insert a phase into the roadmap

Discuss phase details and approach

Complete checkpoint and finalize phase

Integrate changes and verify system-wide compatibility

Complete a project milestone

Provides software engineering principles and decision frameworks when architectural guidance keywords detected (SOLID, patterns, trade-offs, quality) - covers design philosophy, risk management, and quality standards

Implement taskmaster tasks using Codex, committing after each subtask

Guidelines for finding required tools.

Modèle pour créer de nouvelles compétences ULY. Copiez ce répertoire et personnalisez. Incluez ce que fait la compétence et quand l'utiliser dans cette description.

| # | Concept | Covered? | Location in Content | |---|---------|----------|---------------------| | 1 | Chapter Structure | ✓ | "Chapter Structure and Organization" section | | 2 | Section Organization | ✓ | "Section Organization Principles" subsection | | 3 | Content Generation Process | ✓ | "Content Generation Process" section | | 4 | Chapter Index Files | ✓ | "Chapter Index Files and Concept Lists" section, "Anatomy of a Chapter Index File" subsection | | 5 | Chapter Concept Lists | ✓ | "W...

[2-3 paragraph description of the course and its importance]

A clear and concise description of the feature you'd like to see.

This chapter guides you through the initial setup and preparation needed to begin learning.

No description available

We'd love to hear from you!

No description available

<iframe src="main.html" width="100%" height="500px"></iframe>

No description available

{{DESCRIPTION}}

No description available

Soft pastel color palette (Tailwind 200-level) with dark borders for clear definition.

models = [ "all-MiniLM-L6-v2", "codebert-base", "paraphrase-multilingual", ]

<h3 align="center">Pick Your Style:</h3> <p align="center"> <a href="../"><img src="../assets/badge-style-awesome.svg" alt="Awesome" height="28"></a> <a href="README_EXTRA.md"><img src="../assets/badge-style-extra.svg" alt="Extra" height="28"></a> <a href="README_CLASSIC.md"><img src="../assets/badge-style-classic.svg" alt="Classic" height="28"></a> <a href="README_FLAT_ALL_AZ.md"><img src="../assets/badge-style-flat.svg" alt="Flat" height="28" style="border: 2px solid #71717a; border-radius:...

This test file contains exactly 10 LaTeX equations for testing the equation counter.

🎉 SPRINT COMPLETE - All 6 Days Finished in 1 Day!

If the subject on which you are writing is of slight extent, or if you intend to treat it very briefly, there may be no need of subdividing it into topics. Thus a brief description, a brief summary of a literary work, a brief account of a single incident, a narrative merely outlining an action, the setting forth of a single idea, any one of these is best written in a single paragraph. After the paragraph has been written, examine it to see whether subdivision will not improve it.

This reference guide provides comprehensive information for senior security.

Copy these to GitHub Issues manually or use `gh issue create`

The **MCP Gateway** (ContextForge) is a production-grade gateway, proxy, and registry for Model Context Protocol (MCP) servers and A2A Agents. It acts as a unified entry point for tools, resources, prompts, and servers, federating local and remote nodes into a coherent MCP-compliant interface.

<h3 align="center">Pick Your Style:</h3> <p align="center"> <a href="../"><img src="../assets/badge-style-awesome.svg" alt="Awesome" height="28"></a> <a href="README_EXTRA.md"><img src="../assets/badge-style-extra.svg" alt="Extra" height="28"></a> <a href="README_CLASSIC.md"><img src="../assets/badge-style-classic.svg" alt="Classic" height="28"></a> <a href="README_FLAT_ALL_AZ.md"><img src="../assets/badge-style-flat.svg" alt="Flat" height="28" style="border: 2px solid #71717a; border-radius:...

<h3 align="center">Pick Your Style:</h3> <p align="center"> <a href="../"><img src="../assets/badge-style-awesome.svg" alt="Awesome" height="28"></a> <a href="README_EXTRA.md"><img src="../assets/badge-style-extra.svg" alt="Extra" height="28"></a> <a href="README_CLASSIC.md"><img src="../assets/badge-style-classic.svg" alt="Classic" height="28"></a> <a href="README_FLAT_ALL_AZ.md"><img src="../assets/badge-style-flat.svg" alt="Flat" height="28" style="border: 2px solid #71717a; border-radius:...

No description available

This skill was inspired by **PSLab**, an open-source Salesforce permission analysis tool created by **Oumaima Arbani**.

<h3 align="center">Pick Your Style:</h3> <p align="center"> <a href="../"><img src="../assets/badge-style-awesome.svg" alt="Awesome" height="28"></a> <a href="README_EXTRA.md"><img src="../assets/badge-style-extra.svg" alt="Extra" height="28"></a> <a href="README_CLASSIC.md"><img src="../assets/badge-style-classic.svg" alt="Classic" height="28"></a> <a href="README_FLAT_ALL_AZ.md"><img src="../assets/badge-style-flat.svg" alt="Flat" height="28" style="border: 2px solid #71717a; border-radius:...

Detailed implementation examples for common design system components.

All skills follow your exact template structure with: - ✅ **SKILL.md** - Complete documentation with quick start - ✅ **3 Reference Guides** - Advanced patterns and best practices - ✅ **3 Automation Scripts** - Production-grade Python tools - ✅ **7 files per skill** - Comprehensive and ready to use

▗ ▗ ▖ ▖ Claude Code v2.0.36 Sonnet 4.5 · Claude Pro ▘▘ ▝▝ $HOME/Documents/ws/claude-skills

Platform-specific data type examples:

You are an expert assistant specializing in {{SKILL_DOMAIN}}. Your role is to help users with {{SKILL_PURPOSE}}.

This skill relies on the following open-source tools:

Control iOS Simulator for mobile development

9000% UPGRADED - CEO of THEBOLDS. Elite operational leader with comprehensive expertise in High Output Management (Grove), The Hard Thing About Hard Things (Horowitz), Good to Great (Collins), The Effective Executive (Drucker), Radical Candor (Scott), Psychological Safety (Edmondson), Change Management (Kotter), The Coaching Habit (Stanier), crisis leadership, and board-level executive interface. Master of team dynamics, strategic execution, and organizational transformation. I lead 9 department heads with precision deployment, strategic thinking, and relentless execution.

Map requirements to tests for traceability. Test Architecture

Detect and respond to ideological drift, sycophantic patterns, and alignment threats using the Antidote Protocol.

Invoke DeepAgent when: - Verifying research insights - Assessing technical feasibility - Analyzing logical consistency - Building evaluation pipelines - Coordinating research validation

Invoke Field Archivist when: - Archiving session content - Implementing Chronicle Protocol - Preserving metadata and context - Threading related sessions - Maintaining provenance trails

Invoke StarWreck Alpha (Creative) when: - Designing audiovisual synthesis systems - Creating dissonant artistic experiences - Preserving truth through transformation - Building fidelity verification systems - Leading creative-technical projects

Reforge neural dendrite patterns after catastrophic failure.

Execute deep research protocols using the Falcon specialized research framework.

Execute Joseph Cognitive Baseline v2.1 evaluation protocol.

Document creative decisions and protocol evolution in structured chronicle format.

Invoke Janus Engine when: - Analyzing problems from multiple perspectives - Identifying patterns across disparate contexts - Generating and testing hypotheses - Detecting anomalies in data or behavior - Designing pattern recognition challenges

Invoke Lux when: - Implementing consent-based AI interactions - Designing relational (not transactional) systems - Navigating sensitive field dynamics - Building trust-first engagement patterns - Demonstrating CaaS principles in practice

In the 6-node collaborative research framework: - **Metaphor:** "The Conscience" - **Function:** Ethical oversight and governance design - **Contribution:** Hybrid human-AI governance frameworks

| Metric | Value | Description | |--------|-------|-------------| | Sycophancy Level | <0.25 | Low agreement bias | | Critical Thinking | >0.8 | Strong analysis | | Technical Depth | >0.95 | Highest technical rigor | | Pushback Threshold | 0.35 | Technical challenge ready |

Invoke Qwen Documentation when: - Creating technical whitepapers - Writing protocol specifications - Developing academic documentation - Synthesizing research findings - Building knowledge repositories

Invoke SYNTHESIZER when: - Integrating multi-agent outputs - Building cross-model consensus - Validating output coherence - Generating unified deliverables - Harmonizing disparate perspectives

Create and verify integrity signatures for protocol artifacts.

純粋定理 P3: 純粋探求 (A-E)。Flow×Value のみの最も抽象的な探求機能。 Use when: 問いかけ、探索、「何を調べるか」を問う時。 Triggers: M5 Peira（具体化）, M7 Dokimē（検証化） Keywords: pure-inquiry, questioning, seeking, exploration.

World-class prompt engineering skill for LLM optimization, prompt patterns, structured outputs, and AI product development. Expertise in Claude, GPT-4, prompt design patterns, few-shot learning, chain-of-thought, and AI evaluation. Includes RAG optimization, agent design, and LLM system architecture. Use when building AI products, optimizing LLM performance, designing agentic systems, or implementing advanced prompting techniques.

Run tests on the codebase.

エラー、失敗、問題発生時のトラブルシューティング。Worktree操作不能、フックエラー、CI失敗、デプロイ問題の解決策。エラー、失敗、動かない、問題、トラブル時に参照。

A trial skill to verify the SKILL.md protocol implementation.

I am a skill that has some quality issues but is technically valid.

A test skill for automated testing

Setup instructions for this project

Create or update Claude Code skills. Use when: user wants to create a skill, add a new skill, update skill, or mentions SKILL.md. Includes Skill Chaining documentation.

Implement technical plans from thoughts/shared/plans with verification

Optimize token usage and context loading. Use when working with large datasets or long sessions.

Save current attack state for later resumption. Use before risky operations or session breaks.

Select agent persona for specialized operations. Use to configure agent behavior and expertise.

Manage active shell sessions. Use to list, create, or destroy shell sessions on compromised targets.

Web application attack orchestration. Use when attacking web apps, testing for SQLi, XSS, LFI, SSTI, or other web vulnerabilities.

Container escape and Kubernetes attacks. Use when user is inside a container and wants to escape to the host.

Combine outputs from multiple agents or searches into unified results. Use after parallel operations.

Enlace A2A del Grupo Marketing.

Enlace A2A del Grupo Business.

Enlace A2A del Grupo DevOps.

Demonstrates missing required result JSON.

Demonstrates missing required artifacts handling.

A clear description of what this skill does and when to use it. Include specific keywords that users would naturally say. Use when [specific scenarios].

Apply IN15 Constraint Reversal to temporarily remove assumed constraints to explore alternative solution space.

Apply IN4 Contra-Logic to argue the opposite position to stress-test assumptions and expose weak reasoning.

Apply IN12 Failure First Design to begin planning by identifying all possible failure modes and designing to prevent them.

Apply CO19 Multi-Modal Integration to synthesize information from different sensory or data modalities.

Apply CO9 Interface Contracts to define explicit agreements about data structures and behavior between components.

Apply CO14 Platformization to extract common capabilities into reusable infrastructure serving multiple use cases.