Metadata-Version: 2.4
Name: living-ai-framework
Version: 0.1.0
Summary: Persistent Adaptive Intelligence Framework
Author: Charles Washington Juma
License: MIT
Project-URL: Homepage, https://github.com/SAGEWRL/living-ai
Project-URL: Repository, https://github.com/SAGEWRL/living-ai
Project-URL: Documentation, https://github.com/SAGEWRL/living-ai
Keywords: artificial-intelligence,agi,autonomous-agents,memory,reasoning,cognitive-architecture,adaptive-ai,persistent-ai,living-ai
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Operating System :: OS Independent
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: fastapi
Requires-Dist: uvicorn
Requires-Dist: pydantic
Requires-Dist: requests
Requires-Dist: sentence-transformers
Requires-Dist: faiss-cpu
Requires-Dist: numpy
Requires-Dist: torch
Requires-Dist: redis
Requires-Dist: rq
Requires-Dist: psycopg2-binary
Dynamic: license-file

Living AI

Persistent Adaptive Intelligence Framework

«Building the cognitive infrastructure for persistent, adaptive, and autonomous intelligence.»

---

Vision

Living AI is an open cognitive infrastructure framework designed to move artificial intelligence beyond stateless request-response systems toward persistent adaptive intelligence.

Modern AI models are excellent at reasoning within a single conversation but lack long-term continuity. Living AI provides the cognitive architecture required for AI systems to remember, form beliefs, pursue goals, learn from experience, adapt their behavior, and evolve over time.

The long-term vision is to establish a foundation for persistent autonomous intelligence and contribute toward the development of Artificial General Intelligence (AGI).

---

Why Living AI?

Today's AI systems typically suffer from several limitations:

- No persistent identity
- No long-term memory
- No evolving belief system
- No autonomous goal management
- Limited adaptation over time
- No continuous cognitive state

Living AI addresses these challenges by introducing a unified cognitive runtime capable of maintaining persistent intelligence across interactions.

Rather than replacing existing language models, Living AI provides the cognitive layer that enables them to function as adaptive, long-lived intelligent agents.

---

Core Cognitive Architecture

                 User
                   │
                   ▼
        Cognitive Reasoning
                   │
                   ▼
        Semantic Memory System
                   │
                   ▼
          Belief Formation
                   │
                   ▼
          Goal Management
                   │
                   ▼
       Planning & Decision Engine
                   │
                   ▼
      Internal Simulation Engine
                   │
                   ▼
     Reflection & Learning Engine
                   │
                   ▼
      Adaptive Behaviour Engine
                   │
                   ▼
          Natural Response

Every interaction contributes to an evolving cognitive state rather than existing as an isolated request.

---

Current Capabilities

Living AI currently includes:

- Persistent semantic memory
- Redis short-term cognitive memory
- PostgreSQL long-term memory
- Belief extraction and belief evolution
- Goal generation and autonomous goal management
- Cognitive reasoning engine
- Multi-path reasoning
- Decision engine
- Planning engine
- Internal future simulation
- Reflection and self-improvement
- Adaptive behaviour engine
- Emotion modelling
- Cognitive kernel
- Autonomous runtime
- Cognitive observatory support
- Runtime persistence
- Modular architecture

---

Storage Architecture

Living AI separates cognition into short-term and long-term memory.

Redis

Used for:

- Active working memory
- Current cognitive state
- Attention management
- Runtime beliefs
- Active goals
- Runtime context

PostgreSQL

Used for:

- Long-term memory
- Identity persistence
- Goal history
- Belief history
- Reflection history
- Simulation history
- Cognitive evolution

This hybrid architecture enables both fast reasoning and durable persistence.

---

Installation

Clone the repository

git clone https://github.com/SAGEWRL/living-ai.git

Install dependencies

pip install -r requirements.txt

---

Quick Start

Run the interactive agent

python run_ai.py

Example

YOU:
Help me build an autonomous robotics company.

Living AI:
...

---

Running the API

Start the FastAPI server

uvicorn main:app --reload

Available endpoints include:

- GET /
- POST /process
- GET /system/status
- GET /memory/search
- GET /goals
- POST /goals/add
- POST /autonomy/start
- POST /autonomy/stop
- GET /autonomy/status
- GET /autonomy/history

---

Developer Example

from core.system_engine import LivingAISystem

ai = LivingAISystem()

response = ai.process(
    "Help me design an autonomous software company."
)

print(response["response"])

---

Project Mission

To build the foundational cognitive infrastructure that enables AI systems to remember, learn, adapt, evolve, and operate autonomously over long periods of time.

Living AI is designed to become a reusable cognitive framework that developers can integrate into intelligent applications, autonomous agents, robotics, research systems, and future AGI architectures.

---

Roadmap

Phase 1

- Persistent memory
- Belief system
- Goal management
- Reflection engine
- Internal simulation
- Cognitive observatory
- Autonomous runtime

Phase 2

- Python package
- LivingAI Developer SDK
- REST API Platform
- Multi-agent collaboration
- Cloud deployment
- Web dashboard

Phase 3

- LivingAI.ai Platform
- Developer accounts
- API keys
- Hosted cognitive agents
- Enterprise deployment
- Marketplace ecosystem

Long-Term Vision

Create the cognitive infrastructure that enables persistent adaptive intelligence at global scale.

---

Creator

Charles Washington Juma

Founder of the Living AI Framework

---

Contributing

Living AI is under active development.

Contributions, discussions, research, and collaboration are welcome as the framework evolves toward persistent autonomous intelligence.

---

License

See the LICENSE file for licensing information.

---

Project Status

Living AI is an active research and engineering project focused on building persistent cognitive infrastructure for next-generation intelligent systems.

The framework is intended for researchers, developers, startups, and organizations exploring autonomous agents, cognitive architectures, adaptive AI, and the future of Artificial General Intelligence (AGI).
