codemie-tools 0.1.12.2

Package with CodeMie tools for AI Agents

Codemie Tools

Overview

Codemie Tools is a comprehensive toolkit designed to simplify and streamline various development tasks. This toolkit provides a set of tools for access management, notification, code quality, data management, version control, project management, research, PDF processing, and image text extraction.

Installation

To install Codemie Tools, use the following pip command:

pip install codemie-tools

poetry add codemie-tools

For local development, clone the repository and run:

make install

make build

Usage

Using get_toolkit Function

The get_toolkit function is used to get the toolkit instance with the provided configuration.

⚠️ IMPORTANT (v0.1.11.0+): All tool configuration classes that will be stored in the database and displayed in the UI MUST define the credential_type field. This is a mandatory requirement for proper integration with the settings management system. See examples below for the correct pattern.

Here is an example of how to use this function:

from codemie_tools.<toolkit_module> import <ToolkitClass>

# Note: When defining configuration models for database/UI storage,
# ensure they include the credential_type field (see Toolkit sections below)

config = {
    'email': {
        'url': 'smtp.example.com',
        'smtp_username': 'your_username',
        'smtp_password': 'your_password'
    },
    'keycloak': {
        'base_url': 'http://localhost:8080',
        'realm': 'example-realm',
        'client_id': 'example-client-id',
        'client_secret': 'example-client-secret',
        'username': 'admin',
        'password': 'password'
    },
    'sonar_creds': {
        'url': 'http://sonarqube.example.com',
        'sonar_token': 'your_sonar_token',
        'sonar_project_name': 'example_project'
    },
    'elastic': {
        'hosts': ['http://localhost:9200'],
        'username': 'elastic',
        'password': 'password'
    },
    'jira': {
        'url': 'http://jira.example.com',
        'token': 'your_jira_token'
    },
    'confluence': {
        'url': 'http://confluence.example.com',
        'token': 'your_confluence_token'
    },
    'research_config': {
        'google_search_api_key': 'your_google_search_api_key',
        'google_search_cde_id': 'your_google_search_cde_id'
    },
    'root_directory': '/path/to/root_directory',
    'user_id': 'your_user_id',
    'azure_dalle_config': {
        'api_key': 'your_azure_dalle_api_key'
    }
}

toolkit = <ToolkitClass>.get_toolkit(config)
tools = toolkit.get_tools()

Toolkits

Access Management Toolkit

from codemie_tools.access_management.toolkit import AccessManagementToolkit

config = {
    'keycloak': {
        'base_url': 'http://localhost:8080',
        'realm': 'example-realm',
        'client_id': 'example-client-id',
        'client_secret': 'example-client-secret'
    }
}

toolkit = AccessManagementToolkit.get_toolkit(config)

Notification Toolkit

from codemie_tools.notification.toolkit import NotificationToolkit

config = {
    'email': {
        'url': 'smtp.example.com',
        'smtp_username': 'your_username',
        'smtp_password': 'your_password'
    }
}

toolkit = NotificationToolkit.get_toolkit(config)

Code Toolkit

from codemie_tools.code.sonar.toolkit import SonarToolkit

config = {
    'url': 'http://sonarqube.example.com',
    'sonar_token': 'your_sonar_token',
    'sonar_project_name': 'example_project'
}

toolkit = SonarToolkit.get_toolkit(config)

Data Management Toolkit

from codemie_tools.data_management.toolkit import DataManagementToolkit

config = {
    'elastic': {
        'url': ['http://localhost:9200'],
        'api_key': 'elastic'
    }
}

toolkit = DataManagementToolkit.get_toolkit(config)

Version Control Toolkit

from codemie_tools.vcs.toolkit import VcsToolkit

config = {
    'base_url': 'http://gitlab.example.com',
    'access_token': 'your_gitlab_access_token'
}

toolkit = VcsToolkit.get_toolkit(config)

Project Management Toolkit

from codemie_tools.project_management.toolkit import ProjectManagementToolkit

config = {
    'jira': {
        'url': 'http://jira.example.com',
        'token': 'your_jira_token'
    },
    'confluence': {
        'url': 'http://confluence.example.com',
        'token': 'your_confluence_token'
    }
}

toolkit = ProjectManagementToolkit.get_toolkit(config)

Research Toolkit

from codemie_tools.research.toolkit import ResearchToolkit

config = {
    'google_search_api_key': 'your_google_search_api_key',
    'google_search_cde_id': 'your_google_search_cde_id'
}

toolkit = ResearchToolkit.get_toolkit(config)

File System Toolkit

from codemie_tools.data_management.file_system.toolkit import FileSystemToolkit

config = {
    'root_directory': '/path/to/root_directory',
    'user_id': 'your_user_id',
    'azure_dalle_config': {
        'api_key': 'your_azure_dalle_api_key',
        ...
    }
}

toolkit = FileSystemToolkit.get_toolkit(config)

PDF Processing Toolkit

from codemie_tools.pdf.toolkit import PDFToolkit
from langchain_openai import ChatOpenAI

# Create a vision-capable LLM for OCR capabilities
chat_model = ChatOpenAI(model_name="gpt-4-vision-preview", max_tokens=1000)

# PDF processing without OCR
pdf_toolkit = PDFToolkit.get_toolkit(
    configs={"pdf_bytes": pdf_content_bytes}
)

# PDF processing with OCR capabilities for extracting text from images
pdf_toolkit_with_ocr = PDFToolkit.get_toolkit(
    configs={"pdf_bytes": pdf_content_bytes},
    chat_model=chat_model
)

# Get tools for extracting text, analyzing structure, and performing OCR
tools = pdf_toolkit_with_ocr.get_tools()

# Example usage:
pdf_tool = tools[0]

# Get total pages
total_pages = pdf_tool.execute(pages=[], query="Total_Pages")

# Extract text from pages 1-3
text_content = pdf_tool.execute(pages=[1, 2, 3], query="Text")

# Extract text with metadata
text_with_metadata = pdf_tool.execute(pages=[1], query="Text_with_Metadata")

# Extract text from images using OCR
image_text = pdf_tool.execute(pages=[1, 2], query="OCR_Images")

# Use the dedicated OCR tool for more control
ocr_tool = tools[1]  # Available if a chat_model was provided
specific_image_text = ocr_tool.execute(
    pages=[1],  # Process page 1
    image_indices=[0, 1]  # Only process the first two images on the page
)

Configuration Model Requirements (v0.1.11.0+)

⚠️ CRITICAL: All CodeMieToolConfig subclasses that will be stored in the database or displayed in the UI MUST define the credential_type field with the following attributes:

from pydantic import Field
from codemie_tools.base.models import CodeMieToolConfig, CredentialTypes

class MyToolConfig(CodeMieToolConfig):
    """Configuration for my tool."""

    # ✅ REQUIRED: Define credential_type field
    credential_type: CredentialTypes = Field(
        default=CredentialTypes.JIRA,  # Replace with appropriate type
        exclude=True,  # Exclude from serialization
        frozen=True    # Prevent modification
    )

    # ... rest of your configuration fields

Why credential_type is Required

• Database Integration: Used by the backend to categorize and store configurations
• UI Display: Enables proper organization and display of credential settings
• Settings Management: Required for the settings management system to route configurations correctly
• Type Safety: Ensures configuration compatibility with the expected credential type

Available CredentialTypes

Choose the appropriate type from codemie_tools.base.models.CredentialTypes:

• JIRA, CONFLUENCE - Project management
• GIT - Version control (GitLab, GitHub, etc.)
• KUBERNETES, AWS, GCP, AZURE - Cloud platforms
• EMAIL, TELEGRAM - Notifications
• SONAR - Code quality
• ELASTIC, SQL - Data storage
• ZEPHYR_SCALE, ZEPHYR_SQUAD - Test management
• SERVICENOW, REPORT_PORTAL - ITSM and reporting
• And many more (see codemie_tools.base.models.CredentialTypes for full list)

Real-World Example

from pydantic import Field
from codemie_tools.base.models import CodeMieToolConfig, CredentialTypes, RequiredField

class JiraConfig(CodeMieToolConfig):
    """Jira configuration with credential_type."""

    credential_type: CredentialTypes = Field(
        default=CredentialTypes.JIRA,
        exclude=True,
        frozen=True
    )

    url: str = RequiredField(
        description="Jira instance URL",
        json_schema_extra={"placeholder": "https://jira.example.com"}
    )

    token: str = RequiredField(
        description="API token for authentication",
        json_schema_extra={"sensitive": True}
    )

For more details, see the Development Guide.

License

This project is licensed under the MIT License - see the LICENSE file for details.