Agentlify Knowledge Skills

Open-source knowledge skills from Agentlify — the AI agent platform with intelligent model routing.

What Are Knowledge Skills?

Knowledge skills are structured instruction packs that provide AI agents with methodology, checklists, templates, and examples for specific tasks. Unlike traditional tool-calling skills (calculators, web search), knowledge skills teach agents how to think about a problem domain.

Each skill contains:

• When to Use — Clear triggers for when the skill should be invoked
• Workflow — Step-by-step methodology with detailed guidance
• Checklists — Verification items to ensure quality
• Templates — Ready-to-use output formats
• Examples — Concrete input/output demonstrations

How They Work

In Agentlify, agents invoke skills on-demand via get_skill('skill-id') during their workflow steps. This keeps agent prompts lean while giving them access to deep expertise when needed:

Step prompt: "Review this code for security issues."
             "Use get_skill('security-review') for methodology."

→ Agent retrieves the full security review methodology
→ Follows OWASP guidelines, checklists, templates
→ Produces structured security audit

Using These Skills

With Agentlify

Sign up at agentlify.ai to use these skills with intelligent model routing, multi-step agents, and a visual agent builder.

Standalone

Each skill is a self-contained markdown document. You can:

1. Copy into system prompts — Paste the workflow section into your LLM system prompt
2. Use as reference — Follow the methodology manually for your own work
3. Build your own agent system — Parse these files and serve them to your agents
4. Adapt and extend — Fork and customize for your specific needs

With Any LLM

System: You are a code reviewer. Follow this methodology:
[paste code-review.md workflow section]

User: Review this pull request: [code]

Directory Structure

Skills are organized by topic:

skills/
├── README.md
├── development/          # Software development practices
│   ├── code-review.md
│   ├── systematic-debugging.md
│   ├── test-driven-development.md
│   ├── refactoring.md
│   ├── code-execution.md
│   ├── error-handling.md
│   └── performance-optimization.md
├── design/               # System and API design
│   ├── api-design.md
│   ├── agent-workflow-design.md
│   └── database-operations.md
├── content/              # Writing and documentation
│   ├── content-generation.md
│   ├── documentation.md
│   └── output-formatting.md
├── analysis/             # Data and research
│   ├── data-analysis.md
│   ├── research-synthesis.md
│   └── information-extraction.md
├── planning/             # Planning and decisions
│   ├── writing-plans.md
│   └── decision-making.md
├── security/             # Security practices
│   └── security-review.md
├── support/              # Customer support
│   └── support-triage.md
└── quality/              # Quality assurance
    └── self-review.md

Skill Index

Skill	Category	Description
Code Review	Development	Methodology for thorough, constructive code reviews
Systematic Debugging	Development	Root cause analysis and structured debugging
Test-Driven Development	Development	TDD methodology with Red-Green-Refactor cycle
Refactoring	Development	Safely improving code structure
Code Execution	Development	Best practices for running and testing code
Error Handling	Development	Patterns for handling errors gracefully
Performance Optimization	Development	Identifying and fixing performance issues
API Design	Design	RESTful API design best practices
Agent Workflow Design	Design	Designing effective AI agent workflows
Database Operations	Design	Database schema, queries, and migrations
Content Generation	Content	Writing high-quality content
Documentation	Content	Technical documentation best practices
Output Formatting	Content	Structured output and format contracts
Data Analysis	Analysis	Data analysis methodology and reporting
Research Synthesis	Analysis	Multi-step research from provided sources
Information Extraction	Analysis	Extracting structured data from unstructured text
Writing Plans	Planning	Writing clear implementation plans
Decision Making	Planning	Making decisions under uncertainty
Security Review	Security	Security assessment and secure coding
Support Triage	Support	Triaging and handling support requests
Self-Review	Quality	Reviewing and improving outputs before finalizing

Creating Custom Skills

In Agentlify, you can create your own custom knowledge skills with the same structure:

{
  id: 'my-custom-skill',
  title: 'My Custom Skill',
  when_to_use: 'Use this skill when...',
  workflow: '# Workflow\n\n## Step 1...',
  checklists: ['Item 1', 'Item 2'],
  templates: { template_name: 'Template content...' },
  examples: [{ input: '...', output: '...' }]
}

About Agentlify

Agentlify is an AI agent platform that combines:

• Intelligent Model Routing — Automatically selects the best LLM for each task based on cost, quality, latency, and environmental impact
• Multi-Step Agents — Build agents with Research → Plan → Execute → Review workflows
• Knowledge Skills — Equip agents with domain expertise via reusable skill packs
• Visual Agent Builder — Design and customize agents without code

License

These skills are released as open-source educational resources. Use them freely in your projects.

Contributing

Found an improvement? We welcome contributions:

1. Fork this repository
2. Add or improve a skill following the existing structure
3. Submit a pull request

---

Built with ❤️ by the Agentlify team.