Updated 22 June 2026

trovex for Claude Code

Claude Code already reads a CLAUDE.md and rereads your repo to find context. Point it at trovex instead and it gets one canonical doc per query — about 60% fewer tokens.

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. Wire it into Claude Code.

   Start the server, register it over MCP, then run /mcp inside Claude Code to verify trovex shows connected.

   trovex servecopy
   claude mcp add --transport http trovex http://localhost:8765/mcpcopy

   Prefer a file? Add it to a project .mcp.json:

   {
     "mcpServers": {
       "trovex": { "type": "http", "url": "http://localhost:8765/mcp" }
     }
   }

Why Claude Code users want it

A CLAUDE.md is one static file that goes stale and can't route a question to the right doc. trovex keeps many docs canonical and hands Claude Code the single current one that answers — down to the section, with a freshness marker — so it stops rereading files to guess which is the source of truth. About 60% fewer tokens per lookup, and a shared write path so what one Claude Code session learns is read back by the next. See trovex vs CLAUDE.md for the full comparison.