Great AI answers have always depended on one thing: the right information at the right time. When we launched, we cracked that with RAG. Now, we’re taking it to the next level.
We made our agents context-aware.
These agents just know.
They don’t simply respond to a message – they understand the person who sent it. They can see the page the user is on, take in live details you provide (like user’s name or order status), and even explain not only what they know, but how they know it.
Webpage awareness
Your agent can now “read” up to 500 words directly from the exact page it’s embedded on and use that knowledge to its advantage.
This means your customers interact with an agent that uses the article, product, or policy they’re viewing to anticipate their intent and deliver relevant, helpful answers right away, eliminating irrelevant responses and reducing needless back-and-forth.
If you use Live Chat, your agent can now also greet visitors with a highly specific Starter Question tailored to that page – increasing engagement from the very first message.
User information
You can now feed your agent valuable, user-specific details that only you know (first-party data) – the kind of insights that make interactions feel personal and proactive.
With this context, your agent can anticipate needs, offer more relevant solutions, and build stronger connections with your users.
Self-knowledge
We’ve boosted your agent’s awareness of its own knowledge base.
It can now confidently answer questions like “What are you knowledgeable about?” and explain its capabilities with greater certainty – helping users understand exactly how it can assist them.
Don’t forget about the Persona!
The agent’s Persona has always been there, but now it’s more important than ever.
By defining who your agent is and who it works for, you give it a consistent voice and tone that amplifies every other feature, from page-specific knowledge to personalized responses.
Experience Context Awareness Today
Context-Aware Agents represent the next step in intelligent, adaptive AI. They read the page, understand the user, and respond with precision, empathy, and purpose.
Enable Context Awareness now and let your agents truly know your customers.
👉 Enable Context Awareness Now!
Connect with us
Connect with fellow CustomGPT.ai users and exchange experiences on how they made their agents smarter and more proactive! Join our Slack Community to share insights and best practices.
FAQs
What practical problem does “Context Awareness” solve that today’s agents cannot?
Traditional bots rely only on a static knowledge base, so they fall back on generic answers. Our Context‑Aware Agents read the page the visitor is viewing in real time and can accept per‑message facts (order status, user name, etc.). That live context lets them anticipate intent and deliver highly specific, helpful responses—the opposite of “one‑size‑fits‑all” support.
Are these features aimed at support, sales, internal knowledge workers, or all of the above?
All of the above—and more. Any workflow in which understanding what the user is doing right now improves guidance (onboarding, ecommerce, documentation portals, employee help‑desks, etc.) benefits immediately.
How does this differ from simply adding more documents to the knowledge base?
Adding files is retrospective; Context Awareness is prospective and session‑specific. You can’t upload a document that describes the exact paragraph a visitor is reading or the shopping cart they just updated—but the agent can capture that context on the fly and reason over it.
Can I opt in to just one component (e.g., Website Awareness) without enabling the others?
Yes. Website Awareness, Custom Context, and Meta Awareness are independent toggles. Turn on only what fits your use‑case or compliance posture.
Which parts of my site are surfaced to the agent—static HTML, dynamic content, single‑page‑app state, or all?
We crawl the rendered DOM of the page on which the agent sits—up to the first 500 words. Content that requires authentication or client‑side state beyond the public DOM is not captured.
How frequently does the agent re‑read the page if the user scrolls or navigates?
For now, the crawler refreshes once per calendar month per URL. In‑session scrolling does not trigger additional crawls.
Does it respect cookie banners, private pages behind authentication, or user‑specific personalization?
Absolutely. If content is gated behind login, cookies, or personalization tokens, it is not crawled; only publicly accessible markup is ingested.
Will Website Awareness increase latency, and if so, by how much?
Only on the very first request for a new page, while the cache warms. After that, responses are served at normal speed.
Will Website Awareness increase token usage, and if so, by how much?
Yes. Each response that uses page context is billed as the equivalent of two standard agent queries.
What content is actually crawled?
The first 500 words of visible text on the page, excluding hidden elements and script tags.
How many pages per agent can I have?
Up to 1 000 unique URLs per agent are cached simultaneously.
How do I programmatically pass per‑message context—REST, JS SDK, server‑side tag?
Insert your JSON payload into the agent’s embed code as shown in our developer guide (link). REST and server‑side helpers are also available.
Is there a character limit for each injected snippet?
Yes—final value TBA. We’ll publish it in the docs before GA.
What happens if injected context conflicts with the Agent Persona or knowledge base?
The model weighs all inputs and uses best judgement, similar to how it resolves conflicting paragraphs in any prompt. In practice, the most specific, recent detail usually wins.
What types of “meta” questions are supported out of the box?
Anything that asks about the bot—e.g. “What do you know?”, “Where do you get your answers?”, “When was your data last updated?”. The agent detects such intent automatically.
Can I customize or localize the meta responses (“I know about X, Y, Z…”)
Yes. Use the existing Agent Persona field to override tone, language, or disclosure level.
How does the agent decide which knowledge domains it “advertises” if my KB is huge?
If the list of sources exceeds the token budget, the agent selects a representative subset and answers with wording like “including documents such as A, B, C …”.
Is there a risk of disclosing confidential sources when answering meta questions?
No greater than with any other answer. If you prefer zero disclosure, simply disable Meta Awareness in settings.
How is on‑page data processed—client‑side only, streamed to your servers, or cached?
We use a certified third‑party crawler that fetches the public page, caches the text, and passes it to the model. No processing occurs in the visitor’s browser.
Do you store injected custom context, and for how long?
Yes, encrypted at rest, for 30 days to support troubleshooting and analytics, then automatically purged.
What is the expected impact on average response time compared to a vanilla agent?
After the first cached request, there is no additional latency.
How do you handle failure modes—for example, if the website DOM fails to load or an injected context call times out?
The agent falls back to the site’s home‑page cache. If that is unavailable, it proceeds without page context rather than erroring out.
Do you provide monitoring hooks or dashboards for context‑usage metrics?
Yes. The Actions tab shows queries, token counts, and whether page or custom context was used.
Which subscription tiers include Context Awareness?
Premium → Website Awareness + Custom Context
Standard → Meta Awareness
Enterprise plans inherit Premium features plus dedicated SLAs.
Are there overage fees if website context dramatically increases prompt size?
No. Token usage is already accounted for in the “2 × query” pricing noted above.
How do I test the feature in a staging environment before enabling it in production?
Duplicate your production agent, enable the new toggles in the clone, and embed it on a staging URL. No redeployment of production code is required until you’re satisfied.
Does enabling Context Awareness require redeploying existing embed code?
No. All configuration lives in the agent settings dashboard.
Can I chain multiple pages’ contexts (e.g., cart + checkout) into a single conversation?
Yes—pass additional page‑specific facts via Custom Context as the user advances.
Is there an API to fetch the agent’s current “working context” from my backend?
Not yet. It’s on our roadmap and will expose the resolved prompt for observability.
How can I verify that Website Awareness is actually active in real time?
Ask the agent a question that only content on the current page can answer; its response should reference that information verbatim or paraphrased.
Bring context to every conversation.
Enable context-aware agents today and experience AI that truly understands your users, your content, and your goals.
Trusted by thousands of organizations worldwide
