Agentic Development Skills

A collection of agentic development skills for AI-assisted software engineering. Each skill provides structured domain knowledge, reference documentation, and contextual guidance to help AI agents integrate with external services and APIs effectively.

What This Repository Is

This repository contains skills — structured knowledge packages designed for AI agent consumption. Each skill includes:

• A SKILL.md with triggers, descriptions, and core patterns
• Reference documentation covering authentication, workflows, edge cases, and best practices
• Contextual rules and guardrails to guide agent behavior

Skills in this repo are vendor-neutral and designed to work with any AI coding agent.

Repository Layout

.
├── README.md              # This file
├── .gitignore
└── {skill-name}/          # One directory per skill
    ├── SKILL.md           # Skill definition (triggers, description, quick reference)
    └── references/        # Detailed reference documentation
        ├── 01-topic.md
        ├── 02-topic.md
        └── ...

Note: This repository may contain a single skill or multiple skills. Each skill is self-contained in its own directory.

Skills

TaxCloud Integration

Directory: taxcloud/

Integrate the TaxCloud v3 sales tax API into ecommerce applications. Covers real-time tax calculation, order upload, refunds, exemption certificates, TIC codes, address validation, and compliance reporting.

Triggers: TaxCloud, sales tax, nexus, tax compliance, tax quotes, order upload, exemption certificates, TIC codes, filing

Key references:

• Authentication & setup
• Core concepts (nexus, sourcing)
• Real-time cart workflow
• Order upload workflow
• Refunds and credits
• TIC codes and search
• Exemption certificates
• Address validation
• Error handling & rate limits
• TypeScript client generation
• Migration from v1 to v3
• Nexus management
• Integration patterns

Contributing

Adding a New Skill

1. Create a new directory named after the skill (kebab-case or lowercase)
2. Add a SKILL.md following this structure:
   • Frontmatter with name, description, and trigger keywords
   • Overview and links to official documentation
   • Quick-start workflow summaries
   • Key rules and guardrails
   • Endpoint or API quick reference
   • Security considerations
3. Add a references/ subdirectory with detailed documentation
   • Number files sequentially (01-, 02-, etc.) for ordering
   • Cover topics in order of importance / frequency of use
   • Include code examples, error handling, and edge cases
4. Update this README.md to list the new skill

Skill Design Principles

• Agent-first: Write for AI agent consumption, not human tutorials
• Trigger-rich: Include many keywords in the description so agents know when to activate the skill
• Rule-oriented: State hard constraints explicitly ("Never...", "Always...")
• Self-contained: Each skill should work independently without cross-references
• Security-aware: Highlight credential handling, server-side requirements, and common vulnerabilities
• Version-aware: Note API versions, migration paths, and deprecation timelines

License

MIT — or specify your preferred license.