insyt-secure 0.1.3

A service for secure code execution

Insyt Secure

A service that executes Python code snippets received via MQTT messages. Visit https://insyt.co for more information.

Quick Start

Installation

# Basic installation (core functionality only)
pip install insyt-secure

Advanced Installation Options

Insyt Secure provides flexible installation options to include only the dependencies you need:

# Install with PostgreSQL support
pip install "insyt-secure[postgres]"

# Install with MongoDB support
pip install "insyt-secure[mongodb]"

# Install with multiple database support
pip install "insyt-secure[postgres,mongodb,redis]"

# Install with vector database support
pip install "insyt-secure[pinecone]" # or any other vector DB

# Install with cloud provider support
pip install "insyt-secure[aws]" # or azure

# Install with messaging system support
pip install "insyt-secure[kafka]" # or rabbitmq, pulsar

# Complete installation with all dependencies
pip install "insyt-secure[all]"

Available extension categories:

• SQL databases: postgres, mysql, mssql, oracle, clickhouse, snowflake, duckdb
• NoSQL databases: mongodb, redis, cassandra, neo4j, elasticsearch, couchdb
• Vector databases: pinecone, qdrant, milvus, weaviate, chroma, faiss
• Cloud services: aws, azure
• Messaging systems: kafka, pulsar, rabbitmq

Broader categories are also available: rdbms, nosql, vector, cloud, messaging

Basic Usage

# Run with required parameters
insyt-secure --project-id your-project-id-123 --api-key your-api-key-xyz

Getting Help

# View all available options and examples
insyt-secure --help

The help command provides detailed information about all parameters, their defaults, and usage examples directly in your terminal.

Advanced Options

# Run with all options
insyt-secure \
  --project-id your-project-id-123 \
  --api-key your-api-key-xyz \
  --max-workers 10 \
  --execution-timeout 60 \
  --allowed-ips "192.168.1.1,10.0.0.1:3456"

Logging Options

By default, logs are user-friendly and redact sensitive information. You can customize logging behavior:

# Enable more detailed debug logs
insyt-secure --project-id your-project-id --api-key your-api-key --debug

# Show more verbose logs including from third-party libraries
insyt-secure --project-id your-project-id --api-key your-api-key --verbose

# Output logs in JSON format (useful for log processing systems)
insyt-secure --project-id your-project-id --api-key your-api-key --json-logs

# Disable redaction of sensitive information (not recommended for production)
insyt-secure --project-id your-project-id --api-key your-api-key --show-sensitive

You can also control the log level via environment variables:

# Set log level using environment variable
INSYT_LOG_LEVEL=DEBUG insyt-secure --project-id your-project-id --api-key your-api-key

Cross-Platform Compatibility

Insyt Secure is designed to work seamlessly on all major platforms:

• Windows: Fully supported natively, no additional configuration needed
• macOS: Fully supported
• Linux: Fully supported

The service uses paho-mqtt with a platform-agnostic implementation to ensure consistent behavior across all operating systems.

Command Line Parameters

Parameter	Description	Default
--project-id	Your Insyt project ID (required)	-
--api-key	Your Insyt API key (required)	-
--max-workers	Maximum number of concurrent code executions	5
--execution-timeout	Default maximum execution time in seconds per code snippet	30
--allowed-ips	Comma-separated list of allowed IPs/hostnames	All allowed
--verbose	Enable more verbose logs	False
--debug	Enable debug level logging	False
--json-logs	Output logs in JSON format	False
--show-sensitive	Show sensitive information in logs	False

MQTT Credentials Management

This service automatically retrieves and manages MQTT broker credentials:

1. When the service starts, it gets credentials from the Insyt API
2. If the connection drops or credentials expire, it automatically requests new credentials

Use Cases

Data Processing Microservice

Perfect for running data transformation code that connects to various data sources:

insyt-secure --project-id your-project-id --api-key your-api-key --max-workers 15

Secure Environment for Code Testing

Create a sandboxed environment with restricted network access:

insyt-secure --project-id your-project-id --api-key your-api-key \
  --allowed-ips "10.0.0.1,192.168.1.100" --execution-timeout 20

Containerized Deployment

docker run -d --name insyt-secure \
  insyt-secure insyt-secure \
  --project-id your-project-id --api-key your-api-key

System Requirements and Dependencies

Insyt Secure is designed with a modular dependency structure to minimize installation size and resource usage. Below is a breakdown of what's included in each installation option:

Core Dependencies (included in base install)

The base installation includes:

• MQTT connectivity via paho-mqtt
• HTTP client capabilities via httpx
• Basic data science libraries: NumPy, Pandas, SciPy
• JSON and date handling utilities

Optional Dependencies

SQL Database Connectors

• postgres: High-performance PostgreSQL client (asyncpg)
• mysql: MySQL client libraries
• mssql: Microsoft SQL Server connectivity
• oracle: Oracle database connectivity
• clickhouse: ClickHouse analytics database client
• snowflake: Snowflake data warehouse client
• duckdb: Embedded analytical database

NoSQL Database Connectors

• mongodb: MongoDB client with async support
• redis: Redis client
• cassandra: Apache Cassandra and ScyllaDB clients
• neo4j: Neo4j graph database client
• elasticsearch: Elasticsearch search engine client
• couchdb: CouchDB document database client

Vector Database Connectors

• pinecone: Pinecone vector database client
• qdrant: Qdrant vector search engine client
• milvus: Milvus vector database client
• weaviate: Weaviate vector search engine
• chroma: ChromaDB for AI embeddings
• faiss: Facebook AI Similarity Search

Cloud Services

• aws: AWS SDK (boto3) with S3, Dynamo, etc.
• azure: Azure clients for Cosmos DB, Blob Storage, etc.

Messaging Systems

• kafka: Apache Kafka client
• pulsar: Apache Pulsar client
• rabbitmq: RabbitMQ client

Performance Considerations

The base installation already includes the core ML libraries (numpy, pandas, etc.). If you're installing on a resource-constrained environment, consider using only the specific connector extensions you need rather than the broader categories.

For production deployments, we recommend specifying exact dependencies rather than using broader categories:

# Good (minimal dependencies)
pip install "insyt-secure[postgres,redis]"

# Less efficient (pulls in many unused dependencies)
pip install "insyt-secure[rdbms,nosql]"

Platform-Specific Installation

Insyt Secure is designed to work on all major platforms without modification:

Windows

# Install on Windows
pip install insyt-secure

# Run (in PowerShell or Command Prompt)
insyt-secure --project-id your-project-id-123 --api-key your-api-key-xyz

macOS/Linux

# Install on macOS/Linux
pip install insyt-secure

# Run
insyt-secure --project-id your-project-id-123 --api-key your-api-key-xyz

Docker

# Create a simple Dockerfile
echo 'FROM python:3.10-slim
RUN pip install insyt-secure
ENTRYPOINT ["insyt-secure"]' > Dockerfile

# Build the Docker image
docker build -t insyt-secure .

# Run in Docker
docker run insyt-secure --project-id your-project-id-123 --api-key your-api-key-xyz

Platform-Specific Considerations

• Windows: Works natively without WSL, using a cross-platform MQTT implementation
• MacOS/Linux: All features fully supported
• Docker: Ideal for deployment in containerized environments