How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

TokenTracker - Skill - OpenSmartRoute

Name: TokenTracker
Author: DillonSimeone

Imported from DillonSimeone/Website (public/Glue/Daemons/TokenTracker/AGENTS.md). Install upstream with npx skills add DillonSimeone/Website --skill TokenTracker. Copyright stays with the author.

LLM Token Tracker

A Windows system tray application that tracks, aggregates, and visualizes token usage for AntiGravity/Gemini and Cursor. It dynamically discovers local data sources, parses conversation databases, and displays generation stats with optional Cursor dashboard CSV import for billing accuracy.

Features

Multi-Provider Support: Tracks AntiGravity and Cursor in one combined tray icon with per-provider breakdown in the popup.
Dynamic Source Discovery: Automatically finds all ~/.gemini/antigravity* conversation folders and Cursor state.vscdb databases — no hardcoded user paths.
Account-Switch Safe Caching: Unified cache at ~/.llm-token-tracker/cache.json preserves historical usage across account switches.
Delta Caching: Incremental scans by file mtime (AntiGravity) and bubble key tracking (Cursor) for fast refreshes.
Cursor CSV Import: Import dashboard usage-events-*.csv exports for billing-accurate totals; auto-scans ~/.llm-token-tracker/cursor-csv/.
Dynamic System Tray Icon: Color-coded indicator showing today's combined token count.
Sleek Stats Popup: Today/7d/30d/lifetime metrics, per-provider cards, model breakdown, water equivalents.

Project Components

data_sources.py: Dynamic discovery of AntiGravity and Cursor data paths.
models.py: Shared TokenEntry, TokenStats, and aggregation helpers.
antigravity_parser.py: Scans AntiGravity .db / .pb conversation files.
cursor_parser.py: Parses Cursor state.vscdb bubble token data.
cursor_csv_importer.py: Imports Cursor dashboard usage CSV exports.
proto_decoder.py: Pure-Python protobuf wire-format decoder.
token_parser.py: Backward-compatible shim for AntiGravity parsing.
icon_gen.py: PIL-based tray icon generator.
popup_ui.py: Tkinter stats popup.
tray_app.py: Main application lifecycle.
main.py: CLI entry point.

Usage

# Track both providers (default)
python main.py

# AntiGravity only
python main.py --providers antigravity

# Custom data directory
python main.py --data-dir ~/.llm-token-tracker

# Import Cursor CSVs from a folder
python main.py --cursor-csv-dir ~/Downloads/cursor-usage

CLI Options

Flag	Default	Description
`--providers`	`antigravity,cursor`	Comma-separated providers
`--data-dir`	`~/.llm-token-tracker`	Cache and config root
`--cursor-csv-dir`	`~/.llm-token-tracker/cursor-csv`	Auto-import CSV folder
`--cursor-db-max-mb`	`2048`	Warn if Cursor DB exceeds size
`--conversations-dir`	(auto)	Extra AntiGravity conversations path
`--refresh-interval`	`60`	Minutes between auto-refresh
`--debug`	off	Verbose logging

Data Locations

Provider	Discovered Paths
AntiGravity	`~/.gemini/antigravity/conversations/.db`
Cursor	`%APPDATA%/Cursor/User/globalStorage/state.vscdb`
Cache	`~/.llm-token-tracker/cache.json`
CSV import	`~/.llm-token-tracker/cursor-csv/`

Build

pyinstaller LLMTokenTracker.spec

Output: dist/LLMTokenTracker/LLMTokenTracker.exe

Install TokenTracker and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "ocm": "1", "id": "dillonsimeone-website-tokentracker", "kind": "skill", "name": "TokenTracker", "description": "A Windows system tray application that tracks, aggregates, and visualizes token usage for **AntiGravity/Gemini** and **Cursor**. It dynamically discovers local data sources, parses conversation databases, and displays generation stats with optional Cursor dashboard CSV import for billing accuracy.", "publisher": "DillonSimeone", "version": "1.0.0", "capabilities": { "domains": [ "data_analysis", "coding" ], "tags": [ "skill-md", "github-agents" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "A Windows system tray application that tracks, aggregates, and visualizes token usage for **AntiGravity/Gemini** and **Cursor**. It dynamically discovers local data sources, parses conversation databases, and displays generation stats with optional Cursor dashboard CSV import for billing accuracy." ], "primary": false, "metadata": { "source": { "provider": "github-agents", "repository": "https://github.com/DillonSimeone/Website", "path": "public/Glue/Daemons/TokenTracker/AGENTS.md", "ref": "0b6356aca7f8d4f2688ebd8aa220a97c77852d57", "url": "https://github.com/DillonSimeone/Website/blob/0b6356aca7f8d4f2688ebd8aa220a97c77852d57/public/Glue/Daemons/TokenTracker/AGENTS.md", "key": "DillonSimeone/Website/public/Glue/Daemons/TokenTracker/AGENTS.md" } }, "instructions": "# LLM Token Tracker\n\nA Windows system tray application that tracks, aggregates, and visualizes token usage for **AntiGravity/Gemini** and **Cursor**. It dynamically discovers local data sources, parses conversation databases, and displays generation stats with optional Cursor dashboard CSV import for billing accuracy.\n\n## Features\n\n- **Multi-Provider Support**: Tracks AntiGravity and Cursor in one combined tray icon with per-provider breakdown in the popup.\n- **Dynamic Source Discovery**: Automatically finds all `~/.gemini/antigravity*` conversation folders and Cursor `state.vscdb` databases —", "cost": { "context_tokens": 806 } }

TokenTracker

About

LLM Token Tracker

Features

Project Components

Usage

CLI Options

Data Locations

Build

Install TokenTracker and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews