LlamaIndex web search integration

LlamaIndex orchestrates retrieval over your data. ODEN adds the live web to that mix as a tool or a custom retriever — in a few lines, since ODEN is a plain HTTP API.

As a LlamaIndex tool#

import os, requests
from llama_index.core.tools import FunctionTool

def oden_web_search(query: str) -> str:
    """Search the live web and return a synthesized answer with citations."""
    r = requests.post(
        "https://api.oden-api.com/search",
        headers={"Authorization": f"Bearer {os.environ['ODEN_KEY']}"},
        json={"query": query, "include_snippets": True},
        timeout=30,
    )
    r.raise_for_status()
    d = r.json()["results"]
    cites = "\n".join(f"- {c['title']}: {c['url']}" for c in d["citations"])
    return f"{d.get('answer','')}\n\nSources:\n{cites}"

oden_tool = FunctionTool.from_defaults(fn=oden_web_search)
# Pass [oden_tool] to a ReActAgent or FunctionAgent

As a custom retriever#

Return ODEN citations as LlamaIndex nodes so they flow through a query engine:

import os, requests
from llama_index.core.retrievers import BaseRetriever
from llama_index.core.schema import NodeWithScore, TextNode

class OdenRetriever(BaseRetriever):
    def _retrieve(self, query_bundle):
        r = requests.post(
            "https://api.oden-api.com/search",
            headers={"Authorization": f"Bearer {os.environ['ODEN_KEY']}"},
            json={"query": query_bundle.query_str, "include_snippets": True},
            timeout=30,
        )
        r.raise_for_status()
        out = []
        for c in r.json()["results"]["citations"]:
            node = TextNode(text=c.get("snippet", c["title"]),
                            metadata={"title": c["title"], "url": c["url"]})
            out.append(NodeWithScore(node=node, score=c["score"]))
        return out

Notes#

• The tool is best for agents that decide when to search; the retriever is best for query engines that expect nodes.
• score maps straight onto NodeWithScore, so LlamaIndex ranking works out of the box.
• Combine OdenRetriever with your own index retriever for hybrid private + web retrieval.

FAQ#

Can I combine ODEN with a LlamaIndex vector index?#

Yes. Use a router or a composable retriever: your vector index for private data, OdenRetriever for the live web.

Does ODEN return LlamaIndex nodes directly?#

No — it returns JSON. The OdenRetriever above maps that JSON to NodeWithScore in a few lines.

Which is better, tool or retriever?#

Tool for agentic flows where the model chooses to search; retriever for deterministic query engines. Many apps use both.