Framework guide

LlamaIndex

Use Agent-Gantry with query engines, agents, and retrieval workflows. This page covers basic setup and the first production-friendly path; continue to the advanced page for dynamic retrieval, LLM calls, and operations.

Advanced LlamaIndex usage →

Basic page: register once, adapt at the edge

1. Create a Gantry instance near application startup.
2. Register deterministic Python functions with descriptions and input schemas.
3. Export framework-native tools for LlamaIndex immediately before constructing the agent, graph, pipeline, or crew.
4. Keep authorization, rate limits, and audit metadata in Gantry instead of duplicating it inside framework glue code.

from agent_gantry import AgentGantry
from agent_gantry.llamaindex import LlamaIndexAdapter

gantry = AgentGantry()

def get_order_status(order_id: str) -> str:
    """Return the current shipping status for an order."""
    return "in_transit"

gantry.register_tool(get_order_status, tags=["support", "orders"])
adapter = LlamaIndexAdapter(gantry)
framework_tools = await adapter.select("check an order", limit=3)  # Native LlamaIndex adapter export.

First tool-calling loop

Let LlamaIndex own the conversation lifecycle while Gantry owns the tool catalog. Start with a small tool set and assert that the framework receives only safe, documented functions.

Basic LLM call guidance

For early prototypes, use the model client already recommended by LlamaIndex. Wrap that client with a thin policy function that records model name, prompt tokens, completion tokens, tool calls, and latency. Once the integration stabilizes, move the policy into Agent-Gantry's LLM adapter layer so other frameworks share the same rules.