Code Generation "Skills"

Created by Matthew Chartier, Modified on Fri, 22 May at 11:25 AM by Matthew Chartier

Overview

Agiloop IMPLEMENT uses AI-powered code orchestration agents to generate application code for your features.

These agents use the widely adopted SKILL.md format popularized by Anthropic and Claude Code to guide how code should be designed, structured, validated, and refined during generation.

Agiloop includes a set of built-in internal skills designed to produce robust, maintainable, and production-quality code. In addition to these built-in skills, you can also create and manage your own custom skills.


What Are Skills?

Skills are instruction sets that influence how IMPLEMENT generates code.

They allow you to define important implementation guidance such as:

  • Coding standards
  • Architectural preferences
  • Security requirements
  • Compliance requirements
  • Framework conventions
  • Deployment expectations
  • Testing requirements
  • Naming conventions
  • Documentation expectations
  • Validation and review rules

Skills help ensure generated code aligns with your organization’s technical standards and development practices.


Built-In Agiloop Skills

Agiloop IMPLEMENT automatically applies a collection of internal orchestration skills during code generation.

These built-in skills are designed to help produce:

  • Stable application architecture
  • Production-ready implementations
  • Secure coding patterns
  • Strong input validation
  • Consistent project structure
  • Dependency-aware feature generation
  • Deployment-aware output
  • Improved maintainability

In most cases, no additional configuration is required to begin generating high-quality code.


Custom Skills

Custom skills allow you to further tailor the behavior of IMPLEMENT to match your own requirements and preferences.

Organizations commonly create custom skills for:

  • Internal engineering standards
  • Regulatory compliance requirements
  • Preferred frameworks and libraries
  • API design standards
  • Security hardening
  • UI/UX consistency
  • Infrastructure conventions
  • Logging and observability requirements
  • Team-specific workflows

Custom skills can significantly improve consistency across generated features and projects.


Project Skills vs Global Skills

Agiloop supports two types of custom skills:

Project Skills

Project-specific skills apply only to the current project.

Use project skills when:

  • A project has unique technical requirements
  • Different clients require different standards
  • A project uses specialized frameworks or infrastructure
  • You need custom behavior for a single application

Global Skills

Global skills are available across all of your projects.

Use global skills when:

  • You want consistent standards organization-wide
  • You reuse the same architecture patterns
  • Your teams share common coding practices
  • You want reusable compliance or security rules

How to Create or Modify Skills

Step 1: Open Project Settings

Navigate to your project settings.


Step 2: Open the Build Section

Select the Build section.


Step 3: Manage Skills

From the Build section, you can:

  • Create new skills
  • Modify existing skills
  • Enable or disable skills
  • Define project-specific or global scope
  • Refine orchestration instructions used during generation

Changes to skills will affect future code generation operations.


Best Practices

  • Keep skills focused and clearly scoped
  • Avoid conflicting instructions across multiple skills
  • Create reusable global skills for organization-wide standards
  • Use project-specific skills only when needed
  • Include security and validation guidance in all production projects
  • Review and refine skills as your architecture evolves

Example Use Cases

Examples of custom skills include:

  • Enforcing TypeScript strict mode
  • Requiring input validation on all APIs
  • Mandating structured logging
  • Applying company UI component standards
  • Enforcing Kubernetes deployment conventions
  • Requiring security scanning and validation
  • Standardizing API response formats

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article