> ## Documentation Index
> Fetch the complete documentation index at: https://docs.ontoic.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Ask Questions Across Your Knowledge Graph | Ontoic
> Query everything you've captured in plain language. Every answer is grounded in your sources and cited back to specific nodes — zero hallucinations.
Ask lets you query everything you've captured — every answer is grounded in your sources, not training data. Instead of relying on a language model's general knowledge, Ontoic runs retrieval directly across your graph, so every claim traces back to something you've actually read, saved, or researched.
## How to ask
Type any question in natural language into the Ask interface. Ontoic runs graph-native retrieval across your nodes, matching your question against the concepts, documents, and connections you've built up over time.
You don't need to remember exact titles or keywords. Ask in your own words — the same way you'd pose a question to a well-read colleague who has read everything in your graph.
## Reading the answer
Ontoic returns a synthesised answer alongside numbered citations — for example, `[3]` or `[7]`. Each number references a specific node in your graph.
* **Click any citation number** to open the source node directly.
* **From the source node**, follow the link back to the original document, URL, or capture that the node was created from.
Every claim in every answer has a complete chain of provenance: answer → node → original source. Nothing is asserted without a traceable reference.
## When the graph doesn't know
Ontoic never guesses. If your graph doesn't contain enough information to answer your question confidently, it surfaces a **gap** rather than fabricating a response.
A gap appears as a research task you can act on immediately:
* **Start a Dossier** to let Ontoic run an agentic research loop, find external sources, and fill the gap automatically.
* **Capture context manually** by adding notes, uploading documents, or saving URLs relevant to the question.
This means a "no answer" result is genuinely useful — it tells you exactly where your knowledge base needs to grow.
## Query limits
| Plan | Ask queries |
| ---- | ----------- |
| Free | 20 per day |
| Pro | Unlimited |
Ask follow-up questions to go deeper. Within a session, context carries over from one question to the next — you can narrow down, challenge, or extend any answer without starting from scratch.
Go beyond what you've already captured. Dossier runs an agentic research loop to find external sources and fill gaps in your graph.
Learn how Ontoic performs graph-native retrieval and why it produces zero-hallucination answers.