Building an Agent

MCP Integration

Model Context Protocol (MCP) allows agents to connect to external services and data sources through a standardized interface.

from ncp import Agent, MCPConfig

filesystem_server = MCPConfig(
    transport_type="stdio",
    command="mcp-server-filesystem /path/to/files"
)

agent = Agent(
    name="MultiServiceAgent",
    description="Agent with MCP integrations",
    instructions="...",
    tools=[...],
    mcp_servers=[filesystem_server]
)

Data Connector Integration

Agents can query NCP data connectors configured on the platform - such as Splunk, ELK, NetBox, or BigQuery, by referencing them by name.

from ncp import Agent, tool

@tool
def analyze_logs(query: str) -> dict:
    """Query and analyze logs from the connected Splunk instance."""
    ...

agent = Agent(
    name="LogAnalyzer",
    description="Analyzes logs via Splunk connector",
    instructions="...",
    tools=[analyze_logs],
    connectors=["splunk-prod"]
)

Dependency Management

Python dependencies - requirements.txt:

pandas>=1.5.0
requests>=2.28.0

System dependencies - apt-requirements.txt:

curl
wget