> For the complete documentation index, see [llms.txt](https://docs.avonnicomponents.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.avonnicomponents.com/flow/build-with-ai/limitations-and-faq.md). # Limitations & FAQ {% hint style="warning" %} 🚧 **Early access.** Build with AI is rolling out gradually and improving with every release. Something unclear, missing, or broken? [Tell us](/flow/help/support-and-troubleshooting.md): your feedback directly shapes what we improve next. {% endhint %} ## Limitations These tools are focused and deliberately narrow. Knowing where they stop saves you from expecting things they were never built to do. * **Avonni components only, not general flow logic.** The skills create screen flows built around Avonni components and configure those components. They do not build the rest of the flow: decisions, assignments, loops, and record elements stay yours to add in Flow Builder. * **Nothing is deployed.** Every skill writes or edits files locally and stops there. It never pushes to an org. Deploying the flow file and activating the flow is your responsibility, using the Salesforce CLI or your usual deployment process. * **The MCP server is a hard dependency.** The skills refuse to run without it rather than guess. There is no offline mode: if the server is unreachable, the skill stops and asks you to connect it. * **The MCP server documents the latest release, which may be ahead of your installed package.** The documentation service updates automatically with each Avonni package release. If your org is on an older package version, the assistant may confidently use a component or property you don't yet have. Keep your installed Avonni packages up to date to stay in sync. * **Node.js and an authenticated Salesforce CLI are the working foundation, in every client.** The skills run local scripts that require Node.js 18+, and the Salesforce CLI connects them to your org for documentation lookups, component-version queries, and saving your work. Without that foundation, the assistant can read documentation but cannot produce anything usable, including in the Claude desktop app. * **The architect coordinates three artifact skills, not all four.** `avonni-architect` orchestrates the Flow, Dynamic Component, and Experience skills. `avonni-lwc-components` is a direct-edit workflow you invoke on its own. ## Troubleshooting | Problem | Cause | Fix | | ----------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | The skill doesn't activate | The skills aren't installed in the assistant or project you have open. The `npx skills add avonni/skills` command installs into the current project only. | Re-run the install command from the open project, copy the skill folders into your assistant's skills directory (`.claude/skills/` for Claude Code), or add them again from your assistant's Skills settings. | | The skill doesn't activate | The prompt is too vague to trigger the right skill. | Name what you're building explicitly ("Avonni Flow Screen Component", "screen flow"), or invoke the skill by name. In Claude, type `/avonni-flow-components`. | | The assistant guesses attributes instead of looking them up | The MCP server isn't connected. | Check your assistant's MCP status (in Claude Code, run `claude mcp list`) and verify the server URL. Then test with *"List the available Avonni flow components."* | | The skill runs but cannot save or look up org data | Node.js is missing, or the Salesforce CLI isn't installed or isn't authenticated to your org. | Install Node.js 18+, install the Salesforce CLI, then run `sf org login web` and retry. | | The MCP connection fails | Wrong URL, wrong transport, or the server was added to a different project. The `claude mcp add` command registers it for the current project only. | Confirm the URL is `https://mcp.avonnicomponents.com` over HTTP transport. Re-run the command from the right project, or add it at a wider scope (in Claude Code, `claude mcp add --scope user …`). | | The assistant can't find a component I know exists | A `?toolsets=` filter excludes the component's package. With `?toolsets=flow`, dynamic, LWC, and experience components are invisible. | Remove the filter or add the missing package to the list. | | A multi-artifact request only built one piece | The `avonni-architect` skill isn't installed, so a single-artifact skill handled only its own part. | Make sure `avonni-architect` is installed alongside the other skills. | --- # 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, and the optional `goal` query parameter: ``` GET https://docs.avonnicomponents.com/flow/build-with-ai/limitations-and-faq.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. 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.