Setup

The primary way agents drive sootsim is the sootsim CLI. Use the same sootsim path a human uses after the global install; there is no separate agent install.

curl -fsSL https://sootsim.com/install.sh | sh

That’s it. From inside any agent (Claude Code, Cursor, Codex, anything that can run shell commands), sootsim describe, sootsim do tap-text "Sign in", and the rest of the CLI work against whatever sim is running.

#Verifying

With a sim open:

sootsim list           # should show at least one connected tab
sootsim describe       # dumps the current screen

If list is empty, start a sim first (sootsim open <port> or your usual dev flow), then retry.

#Skills

We ship a handful of skill files that teach agents how to use sootsim well: setup, debugging, accessibility, perf, visual regression, and test flow authoring. Drop one into your repo and any compatible agent picks it up automatically.

Two ways to install:

# generic: works in any agent-aware repo
npx skills add tamagui/soot --skill sootsim-debug

# from a soot checkout
sootsim skill install
sootsim skill install codex
sootsim skill install --target .claude/skills sootsim-debug

Available: sootsim-setup, sootsim-debug, sootsim-visual, sootsim-test, and soot. Pick what matches your task, then restart the agent session so it reloads the skill list.

See the skills CLI docs for the generic flow.

#What’s next

• CLI Use: the day-to-day commands an agent runs.
• Agent Devtool (Inspect): the in-sim panel where you talk to an agent and watch it work.