The Knowledge Graph API is the core infrastructure of the Half-Brain, Half-Bot framework.
It enables bidirectional knowledge refinement between AI and humans by maintaining, querying, and evolving structured knowledge graphs.
This API serves as the point of intersection between:
- Human-in-the-Loop (AI → Meat-Brain) → AI guides human learning via structured representations.
- AI-in-the-Loop (Meat-Brain → AI) → Humans refine AI models by correcting hallucinations & enriching knowledge.
By structuring expert knowledge into dynamically evolving graphs, we unlock real-time hypothesis testing and exponential innovation.
Current AI models hallucinate because they lack:
- A stable, structured foundation of knowledge.
- Feedback from domain experts to correct and refine.
Meanwhile, humans are:
- Slow at digesting vast amounts of information.
- Prone to cognitive biases that distort learning.
The Knowledge Graph API solves both problems by:
✅ Storing expert-validated, structured knowledge that AI can use & refine.
✅ Giving humans a graph-based UI to explore & correct AI-generated insights.
✅ Allowing both sides to challenge each other, ensuring models evolve in alignment with empirical reality.
✅ Graph-Based Storage → Supports Neo4j, ArangoDB, and other flexible knowledge graph databases.
✅ Real-Time Graph Querying → Extract structured relationships between concepts, hypotheses, and facts.
✅ Customizable Graph Forking → Experts can fork & edit knowledge graphs to test competing models.
✅ Auto-Generated Graph Visualizations → Browse knowledge structures via interactive UI.
✅ AI-Enhanced Reasoning → AI suggests hypotheses & contradictions for experts to evaluate.
✅ API-First Design → Seamless integration with MeatMentor AI, Half-Brain, Half-Bot, and external AI systems.
git clone https://github.com/Half-Brain-Half-Bot/Knowledge-Graph-API.git
cd Knowledge-Graph-APIpython -m venv venv
source venv/bin/activate # On Windows use: venv\Scripts\activatepip install -r requirements.txtuvicorn app.main:app --reloadThe API will now be available at http://127.0.0.1:8000.
| Method | Endpoint | Description |
|---|---|---|
GET |
/graph/{node_id} |
Retrieve node data from the knowledge graph. |
POST |
/graph/add-node |
Add a new node to the graph. |
POST |
/graph/add-relationship |
Create a relationship between nodes. |
GET |
/graph/query?search={query} |
Search for concepts in the knowledge graph. |
DELETE |
/graph/remove-node/{node_id} |
Remove a node from the graph. |
Full API documentation is available at http://127.0.0.1:8000/docs (Swagger UI).
✅ AI-Assisted Graph Expansion → AI suggests new connections & hypotheses in real time.
✅ Spaced Repetition Integration → Strengthen graph connections via MeatMentor AI.
✅ Decentralized Graph Hosting → Enable peer-to-peer graph synchronization.
✅ Multi-Agent AI Collaboration → Different AI models will challenge & refine each other’s hypotheses.
🚀 Half-Brain, Half-Bot → The parent project integrating AI & human expertise.
🧠 Human-in-the-Loop → Leveraging AI to guide human learning and decision-making through structured knowledge.
🤖 AI-in-the-Loop → Enhancing AI models with expert human insights and curated knowledge.
🔗 Knowledge Graph API → The core API for managing and querying structured knowledge graphs.
Together, these projects create an AI-human synergy where insight and automation reinforce each other.
We’re building a cybernetic intelligence ecosystem.
👥 Contribute: Fork the repo & submit PRs!
💬 Discuss: Open an issue or join our Discord.
🚀 Use It: Start integrating graph-based knowledge into your workflow today!
🧠🤖 Half-Brain, Half-Bot: Individually, we’re useless. Together, we’re slightly less useless.