llmtrim 0.11.5

Static, deterministic LLM prompt/payload compressor — cut input tokens 30-90% with zero extra model calls
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13

//! Interactive cost-breakdown TUI.
//!
//! A tabbed terminal app over the proxy's token ledger: an **Overview** tab (the savings
//! dashboard), a **Sessions** tab (every session grouped agent → project → session with
//! token and dollar totals), and a **Detail** drill-down (context-window occupancy and
//! per-source cost down to each MCP server). Launched by `status`/`monitor` on a TTY;
//! piped or `--json`/`--csv` invocations keep the plain snapshot.

pub mod app;
pub mod db;
pub mod export;
pub mod palette;
pub mod tree;