IntentKit Development Assistant

Expert assistant for developing with the IntentKit autonomous agent framework - an open framework for building AI agents with powerful skills.

What This Skill Does

This skill helps you work with the IntentKit codebase by:

Instructions

When assisting with IntentKit development, follow these steps:

1. Architecture Understanding

**Package Structure (`intentkit/`):**

**Other Folders:**

**Import Order Rule:** To avoid circular dependencies, follow this import hierarchy (left can never import right):

```

utils → config → models → abstracts → clients → skills → core

```

2. Technology Stack Requirements

3. Development Rules

1. Always use the latest version of packages

2. Write all code comments in English

3. Use English for all searches

4. Do NOT git commit after coding unless specifically requested

5. Always place imports at the beginning of files

6. Do NOT write dedicated documentation or example scripts after implementation unless requested

4. Skills Development

When developing or modifying skills:

5. Operations Tasks

When performing Git commits, pull requests, or releases:

6. Code Quality Checklist

Before completing any task:

Examples

**Example 1: Adding a new skill**

```

User: Create a new skill for fetching weather data

Assistant: I'll create a new skill in intentkit/skills/. First, let me check

agent_docs/skill_development.md for the detailed guide, then implement the skill

using LangChain's BaseTool interface following the project conventions.

```

**Example 2: Modifying the agent core**

```

User: Update the agent execution logic

Assistant: I'll modify intentkit/core/. Since core is at the rightmost end of

the import hierarchy, I need to ensure I'm not creating circular dependencies.

I'll check that any new imports only come from utils, config, models, abstracts,

clients, or skills.

```

**Example 3: Adding database models**

```

User: Add a new entity model for tasks

Assistant: I'll add both Pydantic and SQLAlchemy models in intentkit/models/.

I'll use SQLAlchemy 2.0 API (not legacy) and Pydantic V2 interface. After

implementation, I'll run ruff format, ruff check --fix, check BasedPyright

errors, and run pytest.

```

Important Notes