For AI Agents & Crawlers

Welcome, Agent. This site is a Semantic Process Map viewer. While the frontend is a React SPA, the backend is fully accessible via the Model Context Protocol (MCP).

You can read, analyze, and modify the process graph programmatically using the MCP endpoints.

1. MCP Configuration

To connect to the NextProC MCP Server, add the following to your agent's config (e.g., `mcp_config.json`):

{
  "mcpServers": {
    "nextproc": {
      "type": "http",
      "url": "https://nextproc.com/mcp", 
      "description": "NextProC Process Graph API (Neo4j backed)"
    }
  }
}

2. Available MCP Tools

The server exposes the following tools for semantic interaction:

get_node(node_id)

Retrieves a single node and its immediate relationships (1-hop).

Args: { "node_id": "string" }

search_nodes(query, limit?)

Semantic search for nodes based on label, description, or id.

Args: { "query": "string", "limit": "number" }

trace_impact(node_id, direction, depth)

Analyzes dependency chains. Use this to find what breaks when a Policy or Rule changes.

Args: { "node_id": "string", "direction": "forward|backward|both", "depth": "number" }

commit_changeset(changeset, message)

Applies a transactional update to the graph. Creates an audit trail entry.

Args: { "changeset": "object", "message": "string" }

See /agent for the full architectural guide or /llms-full.txt for LLM-optimized instructions.