tokenolo
A local proxy for Claude Code that strips redundant filler from your prompts before they're sent. It runs on your own machine (127.0.0.1) and forwards every request straight to Anthropic with your own key. The only thing it changes is the user text of your prompt, where it removes filler words - deterministically and removal-only. System prompt, tools, tool results, files and history are forwarded untouched, and nothing is stored by tokenolo.
Install
npm install -g tokenolo # or: brew install tokenolo
tokenolo login # free key, no card required
tokenolo wrap claude # launches Claude Code, optimizedAlternative installer: curl -fsSL https://install.tokenolo.com | sh
Integrate
One line, any agent - no new model, no new workflow. Install once, point tokenolo at the agent you already run, and it optimizes every request before it is sent, then gets out of the way.
Use cases
- Privacy-sensitive work - Your prompts and code go straight to Anthropic under your own key; the proxy is local and stores nothing.
- Cache-safe by design - Removal-only, byte-stable edits keep your provider prompt cache valid across turns.
- Zero rewiring - One command wraps the agent you already run - no new model, no workflow change.
Pricing
Every plan has the same features; plans differ only by the daily wrap limit and rate. Annual billing saves two months.
| Plan | Price | Wraps / day | Rate | Notes |
|---|---|---|---|---|
| Pro | $19/mo | 50 wraps/day | 60 req/min | - |
| Team | $49/mo | 500 wraps/day | 120 req/min | - |
| Business | $99/mo | Unlimited wraps/day | 300 req/min | - |
FAQ
- What is it?
- A local proxy for Claude Code that removes redundant filler from your prompts before they're sent. It runs on your machine and forwards your request to Anthropic with your own key.
- Does my code go to tokenolo?
- No. The proxy is local; your files, tool output, and history are forwarded to Anthropic untouched, and tokenolo stores nothing.
- Which agents?
- Claude Code. It's a local middleware layer, not a new model.
- Do I change my workflow?
- No rewiring - same agent, same commands.
- How much does it save?
- Modestly - it removes filler, not context. The value is a privacy-preserving, cache-safe trim with zero workflow change, not a large bill cut.
Links
- Website - https://www.tokenolo.com
- Sign up - https://app.tokenolo.com/signup
- Docs - https://www.tokenolo.com/docs
- Raw markdown - https://www.tokenolo.com/llms.md