#tokens

Get notifications about leaks that might be bleeding your tokens. Based on real-time analysis on your coding agent sessions.

The fastest measured, optimized local Codex usage meter for tokens, sessions, and estimated OpenAI cost.

See where LLM money goes, why, and how to cut waste. Local-first FinOps + spend intelligence for Claude Code & the Anthropic API. Pure stdlib, zero deps.

A Claude Code plugin that shows exactly where your AI coding session wasted tokens — and how to fix it. Flags waste, never quality. Runs locally.

macOS menu bar meter for your Claude Code plan usage — 5h/7d rate-limit windows, account-switch aware, alerts when a saturated quota frees up.

Local-first token usage analytics for coding agents · 你的 token 都花在哪

Context packer for AI Chatbots | No more countless agent tokens needed

Counts input tokens and estimates cost (USD) for OpenAI, Anthropic, and Google.

MemPenny Plus: Claude Memory Manager 2026 - Lean AI Companion Sync & Rollback Tool

Cut your Claude Code bill by 65%. Intelligent context compilation that sends only relevant rules. Zero workflow changes.

Max out your Claude Code & Codex tokens — don't waste the 5-hour quota you paid for. Zero-config CLI: live 5-hour window, token-efficiency score, cost by agent/model/project, cache-waste. Reads local logs, uploads nothing.

Put your Claude Code context on a diet — measure per-session config overhead and find the skills, MCP servers and agents you never use. npx ccdiet

In-flight conversational context compressor and stale-turn pruner to eliminate context rot.

Where did my Claude Code and Codex tokens go? One offline Python script: re-sent history, repeated tool output, files read twice.

Measure what your AI coding sessions actually cost, then act on advice derived from your own numbers — not someone else's.

API-equivalent cost of your Claude Code usage across all your machines, with an archive that outlives transcript cleanup and an offline dashboard

Local token ledger for Claude Code and Codex. Reads the session logs, keeps no paths, writes to SQLite, DuckDB or BigQuery.

Measure the per-turn token tax your AI coding-agent instruction files (CLAUDE.md, AGENTS.md, Cursor, Copilot, Windsurf rules) add to every turn.

Predict rate-limit cutoffs inside your AI agent. Local MCP server + CLI that tracks Claude Max, ChatGPT Plus/Pro/Team, GitHub Copilot, Cursor plan windows and warns before you hit the cap.

Zero-token skill router for Claude Code and Qwen Code. Routes agent requests to the right skill without burning LLM tokens.

Claude Code skill: hand bulk judgment calls to Jev. 86% fewer Claude tokens on a 12-task benchmark, up to 20x faster. One-line npx install.