Purple8 Hyper GraphAI-Native Execution Infrastructure (v0.72.0)

AI Agent & Container Quickstart (60 Seconds) ​

For AI Agents reading this page: Fetch https://docs.graph.purple8.ai/llms.txt or /guides/ai-agent-blueprint for the complete 16-namespace MCP tool specification, parameter contracts, and 5-step application build playbook.

docker run -d \
  --name purple8-graph \
  -p 8000:8000 \
  -e P8G_JWT_SECRET=your-secret-min-32-chars-long-here \
  -e P8G_API_KEY=p8g_dev_key \
  -v purple8-data:/data \
  ghcr.io/purple8-technologies/purple8-graph:developer

# Connect with the focused Starter profile (22 core tools) or omit ?profile= for all 125 tools
claude mcp add --transport http purple8 "http://localhost:8000/mcp?profile=starter" \
  --header "X-API-Key: p8g_dev_key"

{
  "mcpServers": {
    "purple8-graph": {
      "url": "http://localhost:8000/mcp?profile=starter",
      "headers": {
        "X-API-Key": "p8g_dev_key"
      }
    }
  }
}

Build & Query via MCP or Cypher in One Round-Trip ​

Once connected over MCP (http://localhost:8000/mcp?profile=starter), an AI agent can build schemas, ingest documents, orchestrate workflows, and execute hybrid graph + vector queries without writing backend glue code:

# In-process or containerised Python usage:
from purple8_graph import GraphEngine
from purple8_graph.query_language import QueryLanguage
import numpy as np

engine = GraphEngine("./my_graph", auto_config=True)

query_embedding = np.random.rand(384).tolist()
engine.add_node("doc1", labels=["Document"], properties={
    "title": "Supply Chain & Grid Resilience",
    "region": "APAC",
}, embedding=query_embedding)
engine.add_node("alice", labels=["Person"], properties={"name": "Alice Chen"})
engine.add_edge("doc1", "alice", "AUTHORED_BY")

ql = QueryLanguage(engine)
results = ql.execute("""
    CALL db.vector.search('Document', $vec, 10) YIELD node, score
    WHERE node.region = 'APAC' AND score > 0.80
    MATCH (node)-[:AUTHORED_BY]->(author:Person)
    RETURN node.title, author.name, score
    ORDER BY score DESC
""", {"vec": query_embedding})

AI Agent Build Blueprint → · Docker Quickstart → · MCP Integration → · Cypher Reference →

Head-to-Head Architecture Comparison ​

Anyone can theoretically replicate what Purple8 does functionally by stitching together 8 to 12 separate systems, hiring a platform team, and paying a permanent tax in latency, cloud compute, and brittle glue code. Purple8 collapses that entire stack into one engine:

Full benchmark analysis →

License-Based Deployment Across Red Hat, AWS, Azure & GCP ​

Purple8 ships as a hardened container image and self-serve cloud sandbox with license-based capacity tiers: