> For the complete documentation index, see [llms.txt](https://docs.devally.com/EfyQpV4281edZ3h0g8tY/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.devally.com/EfyQpV4281edZ3h0g8tY/integrations/devally-mcp/using-the-mcp.md). # Using the MCP Once the MCP is [connected](/EfyQpV4281edZ3h0g8tY/integrations/devally-mcp/installation.md), you work with DevAlly by asking your AI assistant questions in plain English. This page collects prompts that work well, tips for getting good answers, and the limits of what the MCP will do. *** ## Example prompts Group your questions around the thing you're trying to accomplish. Here are prompts that work well: ### Daily triage * "List all open issues across our workspace, sorted by severity." * "Show me only the critical and serious issues, and group them by product." * "Are there any new issues filed in the last 24 hours?" ### Diving into a specific issue * "Give me everything you have on issue ABC-1 — the WCAG code, the offending HTML, recommended fix, and which pages it affects." * "Show me the before/after code example for case ABC-3." * "Which products have issues against WCAG 2.4.3?" ### Running audits * "Trigger a site audit on the 'Checkout' product." * "Trigger the workflow audit for our login user journey." * "What's the status of the audit I just kicked off?" ### Managing workflows * "List the workflows on our marketing product." * "Show me the steps in the 'Search and download' workflow." ### Reporting * "Build me a summary report of all issues resolved this month." * "How many active issues do we have per product? Format as a table I can paste into Slack." * "Compare issue counts month-over-month for the last three months." *** ## Tips for getting good answers * **Name things explicitly.** "Show me issues on the Checkout product in the Acme workspace" works better than "show me checkout issues." * **Set filters in the question.** Severity, status, and date range help narrow huge result sets — "open critical issues from the last 14 days" beats "show me issues." * **Ask for the format you want.** "Format as a Markdown table" or "give me a Slack-ready summary" saves a follow-up turn. * **Chain follow-ups.** Once you've found an issue, the assistant remembers the IDs and can dig deeper without you re-typing them. * **Trust but verify.** For high-stakes calls (filing tickets, sharing with execs), spot-check the assistant's answer against the DevAlly UI. *** ## What it won't do * It won't make legal or compliance judgments for you — it surfaces WCAG violations and recommendations, not legal advice. * It won't fix code in your repo automatically. It returns the recommended fix; your engineering team applies it. * It won't access anything outside what your DevAlly account is scoped to. Workspaces, products, and audits you can't see in the DevAlly UI won't show up here either. {% hint style="info" %} Not sure what's possible? The MCP is self-documenting — just ask your assistant *"what can you do with DevAlly?"* and it will list the actions available to your account, always up to date. {% endhint %} --- # 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.devally.com/EfyQpV4281edZ3h0g8tY/integrations/devally-mcp/using-the-mcp.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.