Skip to main content

The Open Source Foundation for AI Agents. Powered by the DisCo (Distributed Cognition) architecture.

Project description

Soorma Core SDK

The Open Source Foundation for AI Agents.

Soorma is an agentic infrastructure platform based on the DisCo (Distributed Cognition) architecture. It provides a standardized Control Plane (Registry, Event Bus, Memory Service) for building production-grade multi-agent systems.

PyPI version Python 3.11+ License: MIT

🚧 Status: Day 0 (Pre-Alpha)

Current Version: 0.7.2

We're in active pre-launch refactoring to solidify architecture and APIs before v1.0. The SDK and infrastructure are functional for building multi-agent systems.

Learn more: soorma.ai

Installation

pip install soorma-core

Requirements: Python 3.11+

Quick Start

Note: Infrastructure runs locally via Docker. Clone the repo to get started.

# 1. Clone the repository
git clone https://github.com/soorma-ai/soorma-core.git
cd soorma-core

# 2. Start local infrastructure
soorma dev --build

# 3. Run the Hello World example
cd examples/01-hello-world
python worker.py

# 4. In another terminal, send a request
python client.py Alice

Next steps: See the Examples Guide for a complete learning path.

Core Concepts

Soorma provides three agent types for building distributed AI systems:

  • Worker - Executes domain-specific cognitive tasks
  • Tool - Provides atomic, stateless operations
  • Planner - Orchestrates multi-agent workflows

Platform Services:

  • context.registry - Service discovery
  • context.memory - Distributed state (Semantic, Episodic, Working memory)
  • context.bus - Event choreography
  • context.tracker - Observability

Learn more: See the comprehensive documentation for architecture details, patterns, and API references.

CLI Reference

Command Description
soorma init <name> Create a new agent project
soorma dev Start local infrastructure
soorma dev --build Build and start (first time)
soorma dev --status Show infrastructure status
soorma dev --logs View infrastructure logs
soorma dev --stop Stop infrastructure
soorma dev --stop --clean Stop and remove all data
soorma version Show SDK version

The soorma dev command runs infrastructure (Registry, NATS, Event Service, Memory Service) in Docker while your agent code runs natively on the host for fast iteration and debugging.

Documentation & Resources

📚 Complete Documentation: github.com/soorma-ai/soorma-core

Key Guides:

🎓 Learning Path:

  1. 01-hello-world - Basic Worker pattern
  2. 02-events-simple - Event pub/sub
  3. 03-events-structured - LLM-based event selection
  4. 04-memory-working - Workflow state
  5. 05-memory-semantic - RAG patterns
  6. 06-memory-episodic - Multi-agent chatbot

Contributing & Support

License

MIT License - see LICENSE for details.

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

soorma_core-0.7.2.tar.gz (52.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

soorma_core-0.7.2-py3-none-any.whl (63.4 kB view details)

Uploaded Python 3

File details

Details for the file soorma_core-0.7.2.tar.gz.

File metadata

  • Download URL: soorma_core-0.7.2.tar.gz
  • Upload date:
  • Size: 52.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for soorma_core-0.7.2.tar.gz
Algorithm Hash digest
SHA256 5d5ff59e5450d7858c5667625513774c95571b10a9f3aa7505164952749b0089
MD5 2142c0ad67bb9f920baeb19fed3f6957
BLAKE2b-256 360b104603ebbd463fec2feb436f59c57ed9ba82d3d45c54f9cef9b022300111

See more details on using hashes here.

File details

Details for the file soorma_core-0.7.2-py3-none-any.whl.

File metadata

  • Download URL: soorma_core-0.7.2-py3-none-any.whl
  • Upload date:
  • Size: 63.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for soorma_core-0.7.2-py3-none-any.whl
Algorithm Hash digest
SHA256 8b902ecd9d764058ae633218a42c85a2387d96bb9712c073385b09ece468f4e0
MD5 c39aa06995cdfc4539897d31d7f96d6d
BLAKE2b-256 e3217e57878ff95a584c34cd73ab816db7c0a442662a478eeb48246e9b387a1c

See more details on using hashes here.

Supported by

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