Local proxy that compresses your LLM API requests so you pay less, with no change to the answers. Trims wasted tokens from prompts, history, tool output, and code before they're sent: -31% input / -74% output, measured live. Any provider, no extra model calls. Also an MCP server and embeddable library (Rust, Python, Ruby, Kotlin, Swift).
Drift inferred · capture-to-capture
No drift recorded — single capability capture; advisories appear once its surface changes.
transport stdio counts 0 tools · 0 res
· 0 prompts
permission surface via README inference
no tools enumerated yet for this server.
evidence-backed
findings quoted directly from the published source artifact — not inferred
this server hasn't been statically analyzed yet (npm/PyPI artifacts only).