How to Summarise Intercom Articles with LlamaIndex

import os

import requests
from llama_index.core import VectorStoreIndex
from llama_index.core.schema import TextNode

BASE_URL = "https://api.enconvert.com/v2"
headers = {"X-API-Key": os.environ["ENCONVERT_API_KEY"]}

article_urls = [url.strip() for url in os.environ["INTERCOM_ARTICLE_URLS"].split(",")]  # public Intercom Help Center article URLs

# One synchronous call: no job_id, no polling, no download step
response = requests.post(
    f"{BASE_URL}/distill",
    headers=headers,
    json={
        "urls": article_urls,
        "schema": {
            "title": "The article title",
            "body": "The full article body text",
            "category": "The help centre collection or category the article belongs to",
            "tags": "The article's tags or labels, as a list",
        },
    },
    timeout=120,  # increase for larger batches
)

# Distill is available on paid plans only, and the free tier returns 402
if response.status_code == 402:
    raise RuntimeError("Distill requires a paid plan")
response.raise_for_status()

# Failed pages come back as rows with a failed status, not as errors
nodes = []
for result in response.json()["results"]:
    if result["status"] != "completed":
        print(f"Skipped {result['url']}: {result['status']}")
        continue

    if result.get("render_quality", 1.0) < 0.5:
        # a low score can mean a challenge page or login wall
        print(f"Low render quality for {result['url']}: {result['render_quality']}")
        continue

    data = result.get("data") or {}  # keys match schema exactly: title, body, category, tags
    tags = data.get("tags")
    nodes.append(
        TextNode(
            text=f"{data.get('title')}\n\n{data.get('body')}",
            id_=result["url"],
            metadata={
                "source_url": result["url"],
                "title": data.get("title"),
                "category": data.get("category"),
                "tags": ", ".join(tags) if isinstance(tags, list) else str(tags),
                "render_quality": result.get("render_quality", 1.0),
            },
        )
    )

# Each article is one unit — build nodes directly and skip the default splitter
# LlamaIndex's default embedding model and LLM read OPENAI_API_KEY from the environment
index = VectorStoreIndex(nodes)

query_engine = index.as_query_engine(similarity_top_k=3)
print(query_engine.query("How does a customer change their billing details?"))