Updated 22 June 2026

trovex for Zed

Zed keys MCP servers under context_servers. trovex serves over HTTP, so you bridge it with mcp-remote — and the agent gets one canonical doc per query, about 60% fewer tokens.

{
  "context_servers": {
    "trovex": {
      "command": {
        "path": "npx",
        "args": ["-y", "mcp-remote", "http://localhost:8765/mcp"]
      }
    }
  }
}

trovex is open source and in public beta. The five steps below run in about a minute — no account, no keys.

Quickstart: from install to the tokens saved

1. Install trovex.
   uv tool install trovexcopy
2. Index your repo.

   Embeds your markdown locally (SQLite + ONNX). About a minute on a typical repo.

   trovex index /path/to/repocopy
3. Ask a question.
   trovex search "how do we roll back a deploy?"copy
4. See the tokens saved.

   trovex returns the single current doc that answers — down to the section — and prints how many tokens that saved versus rereading the top few candidates. That number is the point. Look for the green ≈ N tokens saved this query line that confirms it worked.

5. Add it to Zed.

   Start the server, then add a trovex entry under context_servers in Zed's settings.json using the mcp-remote bridge (it proxies trovex's HTTP endpoint into the stdio Zed expects; npx fetches it).

   trovex servecopy

   {
     "context_servers": {
       "trovex": {
         "command": {
           "path": "npx",
           "args": ["-y", "mcp-remote", "http://localhost:8765/mcp"]
         }
       }
     }
   }

Zed's MCP support evolves — if a native HTTP option lands, you can drop the bridge. Check Zed's current MCP docs.

Why Zed users want it

Zed is fast and its agent is sharp on code, but for project knowledge spread across many markdown files it still rereads to find the current one. trovex hands Zed the single canonical doc that answers — section-level, with a freshness marker — so it stops opening stale copies. About 60% fewer tokens per lookup, plus a shared write path across sessions.