> For the complete documentation index, see [llms.txt](https://docs.screendesk.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.screendesk.io/debug-with-ai/overview.md). # AI overview Understand how Screendesk AI turns customer evidence and support knowledge into a clearer next step. Screendesk AI helps your support team understand what happened, find relevant product knowledge, and decide what to do next. It connects the evidence in a customer's recording with the knowledge your team already maintains. There are three parts to the workflow: | Part | What it does | Where you use it | | --------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------- | | **AI & Intelligence** | Analyzes recordings and reports to produce summaries, issue findings, and reproduction steps. Transcripts make spoken content easier to review. | The recording or report in Screendesk | | **Data Sources** | Makes selected documentation, files, procedures, and resolved support conversations searchable as evidence. | The workspace knowledge library | | **AI Resolution** | Investigates an eligible Zendesk ticket using its conversation, recording evidence, published guidance, and trusted knowledge. | A private note in Zendesk, with investigation history in Screendesk | ## From a customer report to a response 1. **Collect the evidence.** Request a Screendesk recording from the helpdesk so the customer can show the problem. 2. **Understand the report.** AI summarizes the recording and identifies useful details, such as observed behavior and steps leading to the issue. 3. **Consult your knowledge.** Data Sources supplies relevant documentation and procedures. Similar resolved conversations can add context. 4. **Investigate the ticket.** For an eligible Zendesk recording, AI Resolution combines that evidence and adds a private investigation note. 5. **Review and respond.** Your agent checks the findings, edits any recommended reply, and sends the response through the helpdesk. For example, a customer records an error while inviting a teammate. The report analysis can identify the action and visible error. Your documentation may explain the required permission. AI Resolution can use both to recommend a next step—or flag the case for human investigation if the evidence is incomplete. ## Start with the features you need * [AI settings](/debug-with-ai/ai-settings.md): enable processing, choose automatic analysis and AI labels, or add your own Gemini key. * [Report intelligence](/debug-with-ai/ai-summary.md): understand summaries, issue findings, and reproduction steps. * [AI transcripts](/debug-with-ai/ai-transcripts.md): review spoken content and jump to moments in a recording. * [Data sources](/debug-with-ai/data-sources.md): connect the knowledge that should support recommendations. * [AI Resolution](/debug-with-ai/ai-resolution.md): configure the Zendesk investigation workflow. * [AI in your helpdesk](/debug-with-ai/ai-in-your-helpdesk.md): see how the capabilities fit each integration. {% hint style="info" %} Feature availability depends on your workspace's plan and enabled features. AI Resolution currently investigates recording-linked Zendesk tickets. Connecting another helpdesk or importing its historical conversations does not enable the same investigation workflow there. {% endhint %} ## Keep knowledge and instructions separate **Data Sources** holds the facts: product documentation, troubleshooting procedures, and support history. **AI Resolution guidance** tells the agent how to work: the evidence it should require, its response style, privacy rules, and when to escalate. Keep product-specific troubleshooting steps in your knowledge library, then use guidance to define how those steps should be applied. ## Use AI alongside your team's judgment An analysis is based on the evidence available in that report. Missing telemetry, unclear audio, or incomplete documentation can limit the result. Review the recording and cited sources when deciding what to tell a customer. AI Resolution posts private notes. It does not send a public reply or mark the customer's problem as solved. ## Hand a technical issue to engineering For a deeper code investigation, connect an AI coding assistant through [Screendesk MCP](/debug-with-ai/mcp.md) or the [Screendesk CLI](/debug-with-ai/cli.md). These give the assistant access to permitted report evidence for debugging. --- # 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.screendesk.io/debug-with-ai/overview.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.