> For the complete documentation index, see [llms.txt](https://help.guestreply.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.guestreply.ai/readme/user-manual/ai-agent/ai-agent-sandbox.md).
# AI Agent Sandbox
The AI Agent Sandbox is a secure testing environment that enables you to validate your AI Agent's responses to guest inquiries without affecting real conversations, before deploying it to production.
**The Sandbox allows you to:**
* **Test AI responses** without affecting real guest conversations
* **Identify missing information** in your knowledge base
* **Validate response quality**
## How to Test Your AI Agent
#### Step 1: Navigate to the Sandbox
Go to [AI Agent → Try](https://app.guestreply.ai/ai-sandbox).
#### Step 2: Select a Testing Scenario
Choose a scenario based on your testing objectives:
**`Lead simulation`** — Evaluates responses to potential guest inquiries.
Use this scenario to test how the AI handles pre-booking questions from potential guests. Simulates conversations with leads who are inquiring about availability, pricing, amenities, etc.
**`Guest simulation`** — Tests responses for confirmed reservations.
Use this scenario to test how the AI communicates with guests who have active bookings. Simulates conversations about check-in procedures, property access, local recommendations, etc.
#### Step 3: Start the Simulation
Click **Start simulation** to begin testing AI responses
## Resolve Knowledge Gaps
When the AI Agent cannot provide an adequate response to a question, you'll see a message indicating that the AI doesn't have enough information to answer. This is your opportunity to identify and fill knowledge gaps in your system.
#### Step 1: Go to the Listing
Click **Go to the listing** to access the knowledge base.
#### Step 2: Add a Field
Click **Add field** to create new knowledge entries.
#### Step 3: Input the Missing Information
Enter the missing information with detailed descriptions. Be specific and comprehensive to ensure the AI can provide accurate responses in the future.
#### Step 4: Re-test
Return to the Sandbox and re-test the same question to validate that the AI can now answer correctly
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://help.guestreply.ai/readme/user-manual/ai-agent/ai-agent-sandbox.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.