Llmtrim MCP Server

Local proxy that compresses LLM API traffic so you pay less. Same answers, smaller bill.

People who work with agentic coding, cost reduction and developer tools and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in Rust.

VERIFIED ACTIVE

LAST COMMIT 2026-09-16 · ★ 235 · #14 OF 278 MAINTAINED API INTEGRATION · VERIFIED 2026-09-18

MPL-2.0 · Rust servers · how we verify → /methodology

01 · Install Llmtrim

before you install - you'll need

Set LLMTRIM_UPSTREAM_PROXY before connecting.

Claude Code

claude mcp add fkiene-llmtrim -- npx -y @llmtrim/cli

Claude Desktop / Cursor / VS Code - add to config

{
  "mcpServers": {
    "fkiene-llmtrim": {
      "command": "npx",
      "args": [
        "-y",
        "@llmtrim/cli"
      ]
    }
  }
}

Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.

Using another client? Same JSON, different key

Claude Desktop · mcpServers

Cursor · mcpServers

VS Code · servers

Windsurf · mcpServers

Zed · context_servers

Cline · mcpServers

Roo Code · mcpServers

Continue · mcpServers

LibreChat · mcpServers

Gemini CLI · mcpServers

Codex CLI · mcp_servers

Full setup guides: every client.

02 · Evidence

Security posture

What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.

runs as local process (stdio) - runs on your machine with your user's permissions

license MPL-2.0 - declared in the repository

npm package @llmtrim/cli - the @llmtrim scope differs from the repo owner (fkiene); confirm it's the project's official package

registry namespace io.github.fkiene is GitHub-verified and matches the repo owner

03 · What Llmtrim can do

Prose above is summarized from the project's README and registry record - no invented capabilities.

Latest releases

v0.13.7 · 2026-09-16

Fixed · DeepSeek dollars follow the published list again. deepseek-flash replaced · deepseek-v4-flash, the retired ids still bill at Flash prices, and a cache hit is · billed at its own $0.003/1M tier instead of a…

v0.13.6 · 2026-09-15

Fixed · Source-file tool results are no longer windowed. Claude Code Read (and · equivalent dumps: guttered cat -n, TypeScript, Rust, Java) was classified as a · log because identifiers such as Exception / Error / info…

v0.13.5 · 2026-09-15

Changed · Interceptor replay/fallback no longer pins blocking threads. Secondary upstream · POSTs (400/422 replay, sub fallback, compact candidates, transport retries) use · the same async HTTP stack as the MITM path…

04 · Who maintains Llmtrim

llmtrim is maintained by fkiene. It's the only MCP server we track from this author; the repo dates to Jun 2026.

05 · Facts

category
API integration - ranked #14 of 278 actively-maintained API integration servers as of 2026-09-18.
release cadence
10+ releases in the last 90 days (latest 2026-09-16)
registry
io.github.fkiene/llmtrim (active, first published 2026-06-22 · 42 versions)
packages
npm:@llmtrim/cli

06 · Llmtrim FAQ

What is Llmtrim?

Local proxy that compresses LLM API traffic so you pay less. Same answers, smaller bill.

Is Llmtrim still maintained?

Yes - as of 2026-09-18, its last commit was 2026-09-16 and it shipped 10+ releases in the last 90 days. We re-verify nightly.

How do I install Llmtrim?

Run `npx -y @llmtrim/cli`. The README documents one environment variable (LLMTRIM_UPSTREAM_PROXY) to set first. Set LLMTRIM_UPSTREAM_PROXY before connecting. You can also paste the ready-made client config above.

Does Llmtrim run locally?

Yes - it's a stdio server: it runs on your machine (via npx) with your user's permissions. Your data stays local unless the server itself calls external APIs.

07 · Alternatives to Llmtrim

More API integration MCP servers · Fr24api MCP · Fatsecret MCP · Foxit PDF API MCP Server · Frisian MCP (Django) · Fundz

More Rust MCP servers · Flash · mwe-MCP - Memory Wiki Engine · Freeq · Docs MCP · Fullbleed PDF Engine · see all