IntentKit Development Assistant

Expert assistant for developing with IntentKit, an autonomous agent framework that enables creation and management of AI agents with capabilities.

What This Skill Does

Provides comprehensive development guidance for the IntentKit autonomous agent framework, including:

Project Architecture

Core Package Structure (`intentkit/`)

Published as a pip package with the following organization:

1. **`intentkit/core/`** - Agent system driven by LangGraph

2. **`intentkit/models/`** - Entity models (Pydantic + SQLAlchemy)

3. **`intentkit/config/`** - System configuration (database, LLM providers, skill providers)

4. **`intentkit/skills/`** - Skills system using LangChain BaseTool for agent capabilities

5. **`intentkit/utils/`** - Utility functions (logging, formatting)

6. **`intentkit/abstracts/`** - Interfaces for core and skills modules

Application Structure (`app/`)

Supporting Directories

Technology Stack

| Component | Technology | Documentation |

|-----------|-----------|---------------|

| Package Manager | uv | - |

| Virtual Environment | .venv | `source .venv/bin/activate` |

| Linting | ruff | Run `ruff format && ruff check --fix` |

| Type Checking | BasedPyright | Ensure no errors in changed files |

| API Framework | FastAPI | https://fastapi.tiangolo.com/ |

| Database ORM | SQLAlchemy 2.0 | https://docs.sqlalchemy.org/en/20/ |

| Data Validation | Pydantic V2 | https://docs.pydantic.dev/latest/ |

| Testing | pytest | Run `pytest` after changes |

Development Rules

1. ✅ Always use the latest package versions

2. ✅ Write all code comments in English

3. ✅ Use English for search queries

4. ❌ Do NOT auto-commit after coding (unless explicitly requested)

5. ✅ Place all imports at the beginning of files

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

Key Guidelines

Avoiding Circular Dependencies

Package import order (left packages can never import from right):

```

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

```

Skills Development

When developing or modifying skills:

Operations & Release Management

When performing Git operations, pull requests, or releases:

Database Modeling

Pydantic Models

Testing Workflow

1. Make code changes

2. Run linting: `ruff format && ruff check --fix`

3. Check types: Ensure no BasedPyright errors in changed files

4. Run tests: `pytest`

Virtual Environment

Always activate before running commands:

```bash

source .venv/bin/activate

```

Typical Workflows

Adding a New Skill

1. Navigate to `intentkit/skills/`

2. Review `agent_docs/skill_development.md`

3. Create skill class inheriting from `BaseTool`

4. Implement required methods

5. Register skill in appropriate provider

6. Write integration tests in `tests/skills/`

7. Run linting and tests

Modifying Agent Core Logic

1. Navigate to `intentkit/core/`

2. Respect dependency order (core is rightmost, can import from all others)

3. Update LangGraph workflows as needed

4. Test changes with integration tests in `tests/core/`

Adding API Endpoints

1. Navigate to `app/` directory

2. Use FastAPI patterns from https://fastapi.tiangolo.com/

3. Follow existing endpoint structure

4. Add tests in `tests/api/`

Database Migrations

1. Modify SQLAlchemy models in `intentkit/models/`

2. Use scripts in `scripts/` for migrations

3. Test with both Pydantic (memory) and SQLAlchemy (storage) representations

Important Notes

Resources