Skip to main content
AI

Local status and usage tracking for Claude Code and Codex

Agent Island monitors local session state for Claude Code, Codex, Cursor, Grok and Antigravity, shows when a run needs your attention, and estimates token-based API value from local records. It is free, open source, and runs on macOS and Windows without requiring an account.

Agenticness = how independently a tool can take action, scored across 9 dimensions. Scored independently by David Kooi, Skylark Creations — see full rubric → Last re-evaluated 11 August 2026.

See pricing ↓

Desktop
API
B2B
For Developers
Usage-Based
CLI
Visit Agent Island

Is this your tool? Claim this listing to manage your content and analytics.

Recent activity

What's happened with Agent Island lately

News mentions sourced from our news feed; score changes from periodic re-evaluations.

Ask about Agent Island

Get answers based on Agent Island's actual documentation

Try asking:

About

What It Is

Agent Island is a desktop companion for coding agents, not a coding agent itself. As of v2.1.2 it supports Claude Code, Codex, Cursor, Grok and Antigravity, and is meant to monitor local session state, surface “your turn” handoffs, and show provider-scoped usage information.

What to Know

Agent Island is useful for tracking and explaining coding-agent usage, but it is not a general-purpose autonomous agent platform — and it deliberately does not act on your behalf, which is what its low agenticness score reflects. Auto-resume was...

Key Features
Monitors local session state for Claude Code and Codex
Shows when a coding run needs your attention
Displays provider-scoped usage surfaces
Reads local Claude Code and Codex usage records
Calculates estimated API value from input, output, cache-write, and cache-read tokens
Use Cases
Track when Claude Code or Codex needs a human handoff during a coding session
Estimate token-based API value from local agent logs for auditing or reporting
Compare provider-reported usage against locally observed session activity
Agenticness: Reactive Tool

Responds to prompts but takes no autonomous action.

High evidence
Last evaluated: Aug 11, 2026

Dimension Breakdown

Action Capability
Autonomy
Planning
Adaptation
State & Memory
Reliability
Interoperability
Safety
Operator Sovereignty

Categories

Pricing
  • Free / open source — full functionality available at no cost.
Details
AddedAugust 3, 2026
RefreshedAugust 14, 2026
Agenticness
Quick Facts
DeploymentHybrid (cloud + self-hosted)
AutonomyCopilot (human-in-loop)
Model supportSingle model
Open sourceYes
Team supportIndividual only
Pricing modelFree / open source
Interfacegui, cli
Sources
Last updated August 23, 2026
Free daily brief

When Agent Island changes its pricing, we catch it the next morning

We recrawl every tool in this directory daily. AI Insider is the 3-minute brief on what actually changed — pricing, features, score moves on our 36-point rubric — often before any press coverage. Sent daily. Skipped when nothing happened.

I use AI for: