Feed Live Web Content into Your LLM's Context

LLMs have training cutoffs. The web doesn't. When your model needs to know about something that happened last week, or needs to read a specific page, denkbot.dog bridges the gap. Fetch any URL, get back clean text, inject it into your prompt. The model reads. You move on.

Get API Key →Read the Docs

What you'd use this for

Augmenting prompts with live documentation, grounding model responses with fresh web content, building ChatGPT-like tools that can browse specific URLs, providing models with current pricing pages, changelogs, or news articles.

How it works

example

import httpx

def prompt_with_url(question: str, url: str) -> str:
    r = httpx.post("https://api.denkbot.dog/scrape",
        headers={"Authorization": f"Bearer {DENKBOT_API_KEY}"},
        json={"url": url, "format": "json"}, timeout=30)
    page = r.json()

    return f"""You are a helpful assistant. Use the following web page content to answer the question.

URL: {page['url']}
Title: {page['title']}

Content:
{page['text'][:6000]}

Question: {question}"""

# Use with any LLM API:
context = prompt_with_url(
    "What are the rate limits?",
    "https://docs.stripe.com/rate-limits"
)