AgentHarbor Blog

Blog

Product updates and engineering write-ups on managing AI coding agents — configs, deploys, usage limits, and what we learned building it.

Featured · 15 Jul 2026

Product

Live usage limits for 17 AI providers, right in your menu bar

Session windows, weekly quotas, credits, monthly spend — every AI provider rations you differently. AgentHarbor tracks all of them locally and warns you before you hit a wall.

AgentHarbor Team

6 min read

Read the post
Engineering

How AgentHarbor keeps your configs safe: backups, drift detection, and undo

Deploying into dotfiles that three other tools also write to is inherently risky. Here is the pipeline that makes every AgentHarbor deploy previewable, reversible, and honest about external changes.

24 Jun 2026 · 7 min read

Product

Introducing AgentHarbor: one harbor for all your AI agents

MCP servers, rules, skills, hooks, and sub-agents scattered across Claude Code, Cursor, Windsurf, and more — plus usage limits you only discover when you hit them. AgentHarbor puts all of it in one native tray app.

10 Jun 2026 · 5 min read