> For the complete documentation index, see [llms.txt](https://docs.datasentinel.io/manual/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.datasentinel.io/manual/ai-connector/test-with-the-sandbox-platform.md). # Test with the sandbox platform Try AI Connector against Datasentinel's public sandbox platform, with shared read-only credentials and no installation required. You don't need your own Datasentinel platform to try AI Connector. A public **sandbox platform** is available at `https://demo.datasentinel.io`, so you can connect an assistant and explore the tools end to end in minutes. It monitors about **20 PostgreSQL instances**, some in [agentless](https://app.gitbook.com/o/DaDQkGCM7r2TghFSBFLQ/s/yE16Xb3IemPxJWydtPOj/features/other-features/agentless-monitoring) mode and the others through a local [agent](https://app.gitbook.com/o/DaDQkGCM7r2TghFSBFLQ/s/yE16Xb3IemPxJWydtPOj/getting-started/installation/agent), so you can try both monitoring modes. {% hint style="info" %} The credentials below are **public by design**: the account is shared and strictly **read-only**. Don't store anything you care about in the sandbox, and expect other people to be using it at the same time. {% endhint %} ## Connection details | | | | -------------- | ---------------------------------- | | Platform | `https://demo.datasentinel.io` | | MCP server URL | `https://demo.datasentinel.io/mcp` | | Username | `sandbox` | | Password | `Mcp-Sandbox-10!` | | Privilege | read-only | ## Connect your assistant Follow any of the configuration guides in this section, using `https://demo.datasentinel.io/mcp` wherever the guide says `https:///mcp`. The sandbox is reachable from the public internet with a valid TLS certificate, so every connection method works, including the Claude desktop app's built-in custom connector. Remember to name the server with the `datasentinel-` prefix, for example `datasentinel-sandbox`, so the Skill discovers it. {% content-ref url="/pages/EHHfMOeh2S0HUizI1dv6" %} [Configure the Claude desktop app for Windows](/manual/ai-connector/connect-a-desktop-app/configure-claude-desktop-windows.md) {% endcontent-ref %} {% content-ref url="/pages/LOr8fRFJ0UAJgJpqrY7c" %} [Configure Claude Code](/manual/ai-connector/connect-a-cli-assistant/configure-claude-code.md) {% endcontent-ref %} {% content-ref url="/pages/tyZchHSlOJ62G7VUaxG8" %} [Connect a compatible assistant](/manual/ai-connector/connect-a-cli-assistant/connect-a-compatible-assistant.md) {% endcontent-ref %} ## Sign in The sandbox MCP server is protected by OAuth 2.1 like any Datasentinel platform. When the assistant asks you to sign in, or when you trigger the sign-in from its MCP settings, the Datasentinel login page opens in your browser: enter the sandbox username and password from the table above. You sign in once; the assistant keeps the session. See [How authentication works](/manual/ai-connector/server-setup/how-authentication-works.md). ## A user token for scripts and the dashboard A script has nobody to sign in, so it authenticates with a **user token** sent as an `Authorization: Bearer` header. Sandbox tokens are valid for **24 hours**; mint a fresh one whenever you need it: ```bash curl -s -u 'sandbox:Mcp-Sandbox-10!' -X POST https://demo.datasentinel.io/ds-api/user-token ``` The response is: ```json { "user-token": "" } ``` Use the value as the header of a scripted client. The live dashboard artifact does not need it: it signs in through the connector. ## Try it out Once the server is connected and the Skill is installed, confirm the wiring with the Skill's built-in checks, then start asking real questions: * `datasentinel platforms` should list `datasentinel-sandbox`. * `datasentinel status` should report a healthy platform with about 20 monitored instances. * Then try a prompt from the examples, such as `Which of the sandbox instances is the busiest right now, and why?` {% content-ref url="/pages/mOz11DkHrYfqd1KeLUgr" %} [Optimize your top queries](/manual/ai-connector/analyze-and-tune-workloads/optimize-top-queries.md) {% endcontent-ref %} ## Publish the live dashboard against the sandbox The sandbox is also the easiest way to try the reference **live dashboard artifact**.
The [`datasentinel-claude-artifact`](https://github.com/datasentinel/datasentinel-claude-artifact) repository holds `datasentinel-postgresql-monitor.html` and the prompt to publish it: add the sandbox as a connector named `datasentinel-sandbox` with the URL `https://demo.datasentinel.io/mcp`, sign in with the sandbox account, then follow the publishing steps in the dashboard guide. No token is needed: the artifact calls the sandbox through your connector. {% content-ref url="/pages/pkOrGusK2I5gkrlD1sWU" %} [Claude Artifact](/manual/ai-connector/dynamic-dashboards/claude-artifact.md) {% endcontent-ref %} --- # 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 by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.datasentinel.io/manual/ai-connector/test-with-the-sandbox-platform.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. 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.