Point your AI tools at my work.
This site runs a Model Context Protocol server. Connect it once and Claude Code, Cursor or any other MCP client can answer questions about my work straight from the source, instead of you reading every page or me sending a PDF.
One command:
claude mcp add --transport http oussama https://oelhaidi.com/.well-known/mcpAdd --scope user to make it available in every project.
Add it to your MCP config file: .cursor/mcp.json for Cursor, .mcp.json for Claude Code, or wherever your client keeps its servers.
{
"mcpServers": {
"oussama": {
"type": "http",
"url": "https://oelhaidi.com/.well-known/mcp"
}
}
}“What did Oussama change in White Label’s booking flow?”
“Which iOS apps has he shipped?”
“Summarise his work history.”
“Is he a fit for a product designer role on a travel product?”
get_profilelist_projectslist_topicsget_topicsearch_siteget_contact_infohttps://oelhaidi.com/.well-known/mcpStreamable HTTP transport, stateless, with no authentication and no API key. POST JSON-RPC to it. A plain GET returns the server card, which lists the same tools as above.
If your tool can't speak MCP, /llms.txt is a plain-text summary of everything here: the work, the roles, the outcomes and the links. Any tool that can fetch a URL can read it. There is also a guide for agents covering every machine-readable endpoint on the site.
Everything the server returns is already public on this site. It is read-only: it can't send me a message, book a call or see anything private. To get in touch, email [email protected] or use the contact page.