The Sandbox lets you test AI responses in a safe environment before deploying configuration changes. Send messages, select which guardrails and knowledge packs to apply, and inspect every detail of the response pipeline — including guardrail evaluations, search results, audit records, and model information.
The Sandbox page displays a configuration panel on the left and a chat panel on the right.
Page Layout
The Sandbox is divided into three sections:
- Configuration panel (left) — select guardrails, knowledge packs, or a group to test with
- Chat panel (right) — send messages and view AI responses
- Results inspector (bottom, collapsible) — inspect guardrail results, search debug data, audit preview, and model information for any response
Configuration
The configuration panel lets you choose between Custom and Group mode.
Configuration Modes
Choose one of two modes at the top of the configuration panel:
| Mode | Description |
|---|---|
| Custom | Select individual guardrails and knowledge packs from dropdown menus |
| Group | Select a pre-configured group — guardrails and knowledge packs are resolved from the group settings |
When you switch modes, selections from the previous mode are cleared automatically.
In Custom mode, two multi-select dropdowns appear:
- Guardrails — choose which guardrails to evaluate against the AI response
- Knowledge Packs — choose which knowledge packs to search for answers
In Group mode, a single dropdown lets you select a group. After selection, the panel displays the guardrails and knowledge packs resolved from that group's configuration.
Context Overrides
Expand the Context Overrides section to simulate different user contexts and, for programmable access work, impersonate a specific credential:
| Override | Options | Purpose |
|---|---|---|
| Channel | Web Chat, Teams, WhatsApp, SMS, Phone, Email, API, CLI, MCP | Simulate any channel, including programmable interfaces |
| Auth Status | Authenticated, Anonymous | Test behavior for authenticated vs. anonymous users |
| Execution Mode | Interactive User, Agent Augmentation | Choose whether responses are rendered for a human or structured for an agent |
| Test as Credential | Any credential from the Programmable Access page | Run the sandbox query as if it came from the selected credential, inheriting its caller type, group, and rate limits |
| Caller Name | Free text | Label the test request with a friendly caller name (for example finance-bot) for audit preview clarity |
The Test as Credential selector is the recommended way to validate new Programmable Access credentials before pointing live traffic at them. It replays the exact identity resolution that the runtime applies to real API, CLI, and MCP requests.
Starting a New Conversation
Click New Conversation to clear all messages and start a fresh thread. This resets the thread context while keeping your configuration intact.
Sending Messages
Type a message in the text field at the bottom of the chat panel and press Enter or click the send button. The Sandbox sends your message to the orchestrator along with your selected configuration.
Each response appears as a chat bubble. User messages display on the right (blue) and assistant responses on the left (gray).
Messages within the same conversation share a thread ID. The orchestrator maintains conversation history across messages, so follow-up questions have full context.
Blocked Responses
If a guardrail tripwire triggers, a yellow warning banner appears above the assistant message. The banner lists which guardrails caused the block.
Inspecting Results
Click the eye icon on any assistant message to open the Results Inspector panel below the chat. The inspector has five tabs:
Guardrails Tab
Shows the evaluation results for each guardrail applied to the response:
- Summary bar — displays how many guardrails passed (for example, "2/3 guardrails passed")
- Per-guardrail cards — each card shows the guardrail name, category, pass/fail status, and expandable reasoning
If a tripwire triggered, an error alert appears at the top listing the blocking guardrails.
Audit Preview Tab
Displays the audit record that would be created in production. Sections include:
| Section | Details |
|---|---|
| Governance | Compliance status (compliant/non-compliant), guardrails checked |
| AI Response | Confidence score with color-coded indicator, response source |
| Escalation | Matched escalation rule (if any), action type, confidence band |
| Metadata | Processing time |
The audit preview also evaluates the response against your tenant's escalation rules in real time.
Search Debug Tab
Shows how the knowledge base search processed your query:
- Original query — what you typed
- Refined query — the query after AI refinement
- Results table — top knowledge base matches ranked by relevance score, with title, content preview, and source
Identity & Access Tab
The Identity & Access tab shows exactly how AskRAI resolved the caller for this request — essential when debugging Programmable Access credentials, role mapping rules, or agent pipelines. The tab is organised into up to four sections:
| Section | Fields |
|---|---|
| Identity | Credential Name, Credential Type, Caller Type |
| Access | Resolved Group (or No group assigned) |
| Context | Channel, Execution Mode |
| Agent Augmentation | Suggested Action (proceed, retry_with_context, or abort), Answer Format, Knowledge Grounded flag, Source Type, and an optional blocked-reason alert |
The Agent Augmentation section only appears when the request was executed in Agent Augmentation mode. Its fields mirror the structured context that the runtime returns to downstream agents, letting you verify the exact payload your agent will receive before shipping.
If the Group field shows No group assigned for a credential you expected to inherit access, check the credential's Group setting in Programmable Access — Personal Tokens inherit from the user's groups instead of carrying a group directly.
Model Info Tab
Displays which AI models processed the request:
| Field | Description |
|---|---|
| Model Tier | The tier level used for this request |
| Refinement Model | Model used for query refinement |
| Guardrail Eval Model | Model used for guardrail evaluation |
| Summarization Model | Model used for response generation |
| Processing Time | Total time in seconds |
If the system is operating in degraded mode (using fallback models due to capacity constraints), a yellow warning appears.
Sandbox conversations are not stored as production audit records. Use the sandbox to test configuration changes before applying them to live groups.
Related Pages
- Guardrails — create and manage the guardrails you test in the sandbox
- Knowledge Base — manage Q&A content that powers search results
- Knowledge Packs — organize knowledge items into collections
- Users, Roles & Groups — configure groups used in Group mode and Caller Type or Credential Type role rules
- Programmable Access — issue the credentials impersonated by the Test as Credential selector
- Settings — set confidence thresholds and escalation rules that affect audit previews