io.github.fkiene/llmtrim
llmtrim
MCP server and proxy that compresses LLM prompts, tool output, and replies to cut token cost.
alive probed 2026-10-15
3tools
746tokens in Claude Opus 5.5
776tokens in Claude Code
0.4 %of a 200k context window
443o200k_base estimate
51 msto initialize
5 msto list tools
- Run it
- npx -y @llmtrim/cli@0.13.4 (stdio, probed in a container with the network off)
- Host
- npmjs.com
- Reports itself as
- llmtrim 0.13.4
- Capabilities
- tools
- Instructions
- 204 characters
- Claude count
- Claude Opus 5.5, measured 2026-10-15. Counted again when the tool definitions change. Method.
- Repository
- https://github.com/fkiene/llmtrim
- Website
- https://github.com/fkiene/llmtrim
- Registry
- active, updated 2026-09-08 · registry entry
- Reachable
- 1 of the last 1 probe
Most expensive tools
| Tool | Tokens (o200k) | Share of this server |
|---|---|---|
| llmtrim_compress | 241 | 54 % |
| llmtrim_compress_text | 138 | 31 % |
| llmtrim_stats | 64 | 14 % |
Badge and feed
[](https://mcpcost.ulehla.dev/s/io.github.fkiene/llmtrim)
An entry for every probe that changed the result, the tool count or the token cost: Atom feed. Add it to a feed reader or a Slack RSS app to hear when the server stops answering or its tool set grows.