Skip to main content

Fiscus is the definitive platform for AI-driven integrations, serving as your AI Integration Engineer.

Project description

Fiscus SDK

The Fiscus SDK is your ultimate AI Integration Engineer. It seamlessly connects AI agents and language models to any API or service, simplifying and automating complex integrations. With Fiscus, you can dynamically orchestrate workflows, execute API operations, and build intelligent AI-powered applications without manual coding.


Features

  • Dynamic API Operations: Integrate with any API or service using connectors, enabling seamless task execution and workflow management.
  • AI Framework Integration: Easily work with frameworks like LangChain, Crew, and OpenAI Function Calling.
  • Flexible Data Handling: Retrieve responses in plain text or JSON format based on your configuration.
  • Comprehensive Callbacks: Manage success, errors, authentication, streaming, logging, and debugging with ease.
  • Secure User Management: Handle tokens, OAuth flows, and RBAC to support multiple users and services.
  • Effortless Connector Expansion: Add new connectors without manual integration, empowering you to scale rapidly.
  • Advanced Orchestration: Automatically route and manage workflows across unlimited connectors for scalable operations.
  • Robust Error Handling: Handle issues dynamically with pre-built hooks and authentication retries.

Installation

Install the Fiscus SDK via pip:

pip install fiscus-sdk

System Requirements

  • Operating Systems: Windows, macOS, Linux
  • Python Versions: 3.9+

Quick Start

Example: Sending an Email

Stateless Backend

from fiscus import FiscusClient

# Initialize the Fiscus client
client = FiscusClient(
	api_key='YOUR_FISCUS_API_KEY',
	user_id='user_123'
)

# Define email parameters
email_params = {
	"emailData": {
		"headers": [
			{"name": "To", "value": "example@test.com"},
			{"name": "From", "value": "support@fiscusflows.com"},
			{"name": "Subject", "value": "Test Email"}
		],
		"body": {
			"mimeType": "text/plain",
			"data": "Hello, this is a test email sent via API."
		}
	}
}

# Execute the email operation
response = client.execute(
	connector_name='gmail',
	operation='send_email',
	params=email_params
)

# Check response
if response.success:
    print('Email sent successfully!')
else:
    print(f'Error: {response.error_message}')

Stateful Backend

from fiscus import FiscusClient, FiscusUser

# Initialize the Fiscus client
client = FiscusClient(api_key='YOUR_FISCUS_API_KEY')

# Create a user instance
user = FiscusUser(user_id='user_456', client=client)

# Execute the email operation
response = client.execute(
	connector_name='gmail',
	operation='send_email',
	params=email_params
)

# Check response
if response.success:
    print('Email sent successfully!')
else:
    print(f'Error: {response.error_message}')

Core Concepts

  • FiscusClient: The main entry point for the SDK. Create and manage users, and execute operations.
  • FiscusUser: Represents a user, managing user-specific authentication, preferences, and RBAC.
  • FiscusConnector: Manages integrations with external APIs or services.
  • Operation: A specific function provided by a connector, like sending an email.
  • Callbacks: Customizable hooks for logging, debugging, and handling responses.
  • Dynamic Preferences: Customize behavior per user for seamless API operations.

AI-Driven Workflows

Using ai_execute for Adaptive Automation

The ai_execute function dynamically interprets natural language instructions to execute workflows using connectors and language models. It supports sequential and conditional execution, making it the backbone for intelligent task automation.

Benefits

  • Natural Language to Workflow: Converts user input into actionable tasks.
  • Multi-LLM Compatibility: Supports OpenAI, Anthropic, and others.
  • Memory Management: Stateful and stateless configurations.
  • Flexible Execution: Use callbacks and custom decision logic for complex workflows.

Example

from fiscus import FiscusClient, FiscusUser, FiscusLLMType

# Initialize Fiscus client and user
client = FiscusClient(api_key='YOUR_FISCUS_API_KEY')
user = FiscusUser(user_id='user_123', client=client)

# Execute AI workflow
response = client.ai_execute(
    input="Schedule a meeting and email participants.",
    llm_type=FiscusLLMType.OPENAI,
    user=user
)

if response.success:
    print("Workflow executed:", response.result)
else:
    print(f"Error: {response.error_message}")

Troubleshooting

Common Installation Issues

  • Permission Errors: Use sudo or a virtual environment.
  • Proxy Issues: Configure pip to use proxy settings.

For additional help, contact support@fiscusflows.com.


License

The Fiscus SDK is proprietary software developed by Fiscus Flows, Inc. By using this SDK, you agree to the terms and conditions outlined in the LICENSE file included with the distribution. For any licensing questions, contact support@fiscusflows.com.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fiscus-0.1.0a8.tar.gz (40.4 MB view details)

Uploaded Source

Built Distribution

fiscus-0.1.0a8-py3-none-any.whl (54.7 kB view details)

Uploaded Python 3

File details

Details for the file fiscus-0.1.0a8.tar.gz.

File metadata

  • Download URL: fiscus-0.1.0a8.tar.gz
  • Upload date:
  • Size: 40.4 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.12.7

File hashes

Hashes for fiscus-0.1.0a8.tar.gz
Algorithm Hash digest
SHA256 b1a7e6515145b73ffa201bf5429a14070a859f8577ec4baf3dc646e1dea76651
MD5 5b96d2ba359ca28b62dff39a1c2ee1c2
BLAKE2b-256 e736cb76240e3aece93bdc41b18ef250d03327204ce2f6b8ddb7f0a99ad1b804

See more details on using hashes here.

File details

Details for the file fiscus-0.1.0a8-py3-none-any.whl.

File metadata

  • Download URL: fiscus-0.1.0a8-py3-none-any.whl
  • Upload date:
  • Size: 54.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.12.7

File hashes

Hashes for fiscus-0.1.0a8-py3-none-any.whl
Algorithm Hash digest
SHA256 3819571821a08602cc4f2d05035cbacb21d28a8dce02e20c5e557b06923ab53f
MD5 dae30edc1f9fc4375157bef1ac128ef3
BLAKE2b-256 88f663c57bf0d5e21b232f93199ca0b26647130a67dbcb581fbadf312846973e

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page