About crystl

Introduction

I used Claude Code in a plain terminal for about a year before I started building crystl. As an agency owner, I often run agents across half a dozen projects in a day. As I started doing more with them, I kept running into problems I couldn't easily resolve.

I tried the other apps I could find. Some I really liked, but none gave me the control and options I wanted. So I started building a terminal around the way I was working.

What I kept running into

I was switching tabs constantly to find out which agent needed me. Floating cards and the activity panel show me who's working, who's stuck, and what they need to keep going.

I wanted my project history kept on my own computer. Conversation history lets me search past sessions and see what my agents said and ran.

I wanted agents to start with the project's context, without writing a long catch-up prompt every time. Project instructions give them that context. Heroes let me define specialists with their own models and instructions for particular kinds of work.

I wanted one agent to see what another was doing. The terminal keeps structured session data, so agents can read each other's screens and check on work in progress.

I wanted several agents working on the same project without overwriting each other's edits. Isolated shards give each agent its own git worktree and branch, keeping their edits separate until it's time to bring them together.

I was digging through notes to find prompts and commands I'd already written. Now I can save them as facet inserts and use them with a click.

I wanted to decide how agents divided up the work and who reported to whom. With multi-tiered orchestration, I can choose the models, write the instructions, and watch the agents work in their own sessions.

I wanted work to happen while I was away from the desk. Schedules launch it at the right time. Sequences run steps in order: one agent builds, another reviews, and another implements fixes. The iPhone app lets me check in when I'm not home.

You can ask your agent

An agent launched inside crystl gets access to the app's documentation and a CLI it can use to operate it. You can ask it to set up a schedule, start agents on separate branches, or check on another session. You don't have to find every feature yourself before you can use it.

What it is

crystl is a Metal-accelerated macOS terminal built for multitasking with AI agents. It's like an operating system for your projects. crystl is accessible for beginners and a power tool for advanced workflows. Everything printed to the terminal in crystl is a block with unique metadata. Every message, tool call, and code change is tracked and tagged, so your agents can see what's happening in real time. It also makes them time and date aware.

Claude Code, Codex, and Antigravity CLI are first-class agents (any CLI agent runs), and it functions as a normal terminal too. crystl organizes projects into gems, manages terminal sessions as shards, and handles permission approvals through floating glass panels without hunting for tabs or windows. When you need to run multiple agents on the same repo, the isolated shards feature creates a separate git worktree and branch for each agent, keeping their live in-repo edits apart until integration.

crystl quest lets you build agent teams with their own skills. They talk to you and to each other, so they get work done faster and with fewer mistakes.

With crystl, every conversation is preserved. You can come back hours or days later and see exactly what was accomplished, what was approved, and the context of each session.