Install
The server is remote: there is nothing to download. Add this URL to your AI client as a remote MCP server:
https://agents.avineon.solutions/mcp
Transport: Streamable HTTP. Authentication: none. Menus and plans change, so each section links to the vendor's own instructions (checked 14 November 2026).
Jump to: Claude · Claude Code · ChatGPT · VS Code · Cursor · Gemini CLI · Other clients · Check the connection · Troubleshooting
Claude (web and desktop)
- Go to Customize → Connectors.
- Select + Add, then Add custom connector.
- Enter a name, such as Avineon, and the URL
https://agents.avineon.solutions/mcp. Select Continue. - Under Authentication, choose No sign in, then select Add.
- In a chat, select +, then Connectors, and turn Avineon on.
Free plans can add one custom connector. On Team and Enterprise plans, an owner adds it under Organization settings → Connectors, and members then select Connect under Customize → Connectors. Claude connects from Anthropic's cloud, also in the desktop app. Source: Claude Help Center.
Claude Code
claude mcp add --transport http avineon https://agents.avineon.solutions/mcp
Add --scope user to use it in every project, or --scope project to share it with your team through the project's .mcp.json:
{
"mcpServers": {
"avineon": {
"type": "http",
"url": "https://agents.avineon.solutions/mcp"
}
}
}
Check it with claude mcp list, or /mcp inside a session. Source: Claude Code documentation.
ChatGPT
- In ChatGPT on the web, go to Plugins.
- Select the plus button, then Add custom MCP server.
- Enter a name, and under Connection the URL
https://agents.avineon.solutions/mcp. - For authentication, choose No authentication.
- Read the warning, tick the box to continue, and select Create. Install the plugin, then type
@in a chat to select it.
OpenAI lists custom MCP servers for ChatGPT Business, Enterprise and Edu on the web, where admins may control who can add them, and for Pro users with read and fetch permissions. Source: OpenAI Help Center.
VS Code (GitHub Copilot)
Run MCP: Add Server from the Command Palette, choose HTTP, and enter the URL https://agents.avineon.solutions/mcp and the ID avineon. Or add a .mcp.json file at the root of the workspace:
{
"mcpServers": {
"avineon": {
"type": "http",
"url": "https://agents.avineon.solutions/mcp",
"tools": [
"*"
]
}
}
}
Older set-ups use .vscode/mcp.json, with a servers object in place of mcpServers. Source: VS Code documentation.
Cursor
Add this to .cursor/mcp.json in your project, or to ~/.cursor/mcp.json to use it everywhere:
{
"mcpServers": {
"avineon": {
"url": "https://agents.avineon.solutions/mcp"
}
}
}
Source: Cursor documentation.
Gemini CLI
gemini mcp add --transport http avineon https://agents.avineon.solutions/mcp
Or add it to .gemini/settings.json in your project, or ~/.gemini/settings.json. Use httpUrl, which is Streamable HTTP; url means the older SSE transport:
{
"mcpServers": {
"avineon": {
"httpUrl": "https://agents.avineon.solutions/mcp"
}
}
}
Check it with gemini mcp list, or /mcp inside a session. Source: Gemini CLI documentation.
Other clients
Any client that supports remote MCP servers over Streamable HTTP works. Add a remote (HTTP) server with the URL https://agents.avineon.solutions/mcp and no authentication. To try the server without an AI client, run npx @modelcontextprotocol/inspector and connect to the URL with the Streamable HTTP transport.
Check the connection
Paste this into a new chat:
Is the Avineon MCP server connected? List every tool it offers, with one line on what each does. If you cannot see its tools, say so rather than answering from memory.
A correct answer lists 18 tools, including find_transformers_for_task, find_fme_training and estimate_fme_flow_hosted_cost. Some clients show the names with a prefix, such as mcp__avineon__find_fme_training in Claude Code.
All 18 tools
get_company_overviewlist_servicesget_pagesearch_sitesearch_case_studiesget_contact_infolist_eventssearch_fme_transformersget_transformer_detailslist_transformer_categoriesfind_transformers_for_tasksearch_fme_docsfind_fme_trainingget_arcgis_online_backup_managerestimate_arcgis_online_backup_costrecommend_fme_licenceget_fme_licensing_overviewestimate_fme_flow_hosted_cost
Troubleshooting
- The server is not listed
- Check that the URL ends in
/mcpand has no other port or path. Restart the client after editing a configuration file. In Claude Code, a server added without--scopeis only available in the project where you added it. - Listed, but disconnected or failing
- Open https://agents.avineon.solutions/health in a browser: it should show
"status":"ok". If it does, check that the client is set to Streamable HTTP, not SSE or a local command, and that no proxy or firewall on your network blocks the address. Claude and ChatGPT connect from their own cloud, so your network is not the cause there. - Connected, but no tools
- Some clients need the server turned on for each chat: in Claude, + → Connectors; in ChatGPT, type
@and select it; in VS Code, check the tools picker in Copilot Chat. Then ask again in a new chat. - Rate limited (HTTP 429)
- The server allows 120 requests a minute per client. Above that it replies with HTTP 429, a
Retry-After: 60header and the message "Rate limit exceeded; try again in a minute." Wait a minute and retry.