vector-knowledge-graph-mcp
v1.0.0
Published
AI-powered vector knowledge graph MCP server for agents. Supports add node, add edge, semantic node search. By MEOK AI Labs.
Downloads
54
Readme
Vector Knowledge Graph
By MEOK AI Labs — MEOK AI Labs MCP Server
Vector Knowledge Graph MCP Server — Neo4j-style graph + vector hybrid for compliance reasoning.
Installation
pip install vector-knowledge-graph-mcpUsage
# Run standalone
python server.py
# Or via MCP
mcp install vector-knowledge-graph-mcpTools
add_node
Add a node to the knowledge graph with properties, embeddings, and metadata.
Parameters:
label(str)properties(str)node_id(str)
add_edge
Create a directed edge between two nodes with relationship type and weight.
Parameters:
from_id(str)to_id(str)relation(str)weight(float)
semantic_node_search
Search for nodes using semantic similarity matching against stored embeddings.
Parameters:
query(str)top_k(int)
trace_compliance_chain
Trace the compliance chain from a requirement through controls to evidence.
Parameters:
start_node_id(str)max_depth(int)
find_gaps
Find gaps in the knowledge graph where expected relationships or nodes are missing.
Parameters:
required_frameworks(str)
Authentication
Free tier: 15 calls/day. Upgrade at meok.ai/pricing for unlimited access.
Links
- Website: meok.ai
- GitHub: CSOAI-ORG/vector-knowledge-graph-mcp
- PyPI: pypi.org/project/vector-knowledge-graph-mcp
License
MIT — MEOK AI Labs
