How do I install the Zstack CLI skill?

Run npx skillmds@latest add idanbeck/zstack-cli in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Zstack CLI safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Zstack CLI?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Zstack CLI free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Zstack CLI?

idanbeck (@idanbeck) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Zstack CLI by idanbeck — Agent Skill

Name: Zstack CLI
Rating: 4.5 (4 reviews)
Author: idanbeck

Zstack CLI

Use when operating ZStack through its first-party `zstack` CLI, including logging in, creating API tokens, listing workspaces, deploying environments, streaming logs, wiring domains, managing resources/units, or opening runtime shells for ZStack-backed apps.

idanbeck Updated Sep 9, 2026

File contents

ZStack CLI

Use the zstack CLI instead of scraping the ZStack web UI whenever you need to inspect or operate workspaces, deployments, resources, domains, deploy logs, runtime DB/files/secrets, admin capacity, GitHub source discovery, or shell access.

Defaults

Run from /Users/idanbeck/zerg-stack/zstack unless the user asks for another repo.
CLI source lives in cli/zstack.
Config is ~/.config/zstack/config.json; environment overrides are ZSTACK_BASE_URL, ZSTACK_TOKEN, ZSTACK_SESSION_TOKEN, ZSTACK_WORKSPACE, ZSTACK_CONFIG_PATH.
Prefer API tokens for automation. Browser session login is fine for local operator tasks.
Prefer JSON-returning commands for scripts and agents.

Build

cd /Users/idanbeck/zerg-stack/zstack/cli/zstack
go build -o zstack .

Use ./zstack ... from the build directory or install the binary onto PATH.

Common Workflows

Authenticate:

zstack auth login --email "$EMAIL" --password "$PASSWORD"
zstack auth whoami

Create a workspace-bound automation token:

zstack auth token create \
  --name agent \
  --workspace zstack \
  --scope workspace:write \
  --scope deploy:write \
  --scope resource:write \
  --scope domain:write

Inspect workspaces and deployments:

zstack workspace list
zstack workspace show zstack --json
zstack workspace branches --workspace zstack
zstack workspace branch-deploy --workspace zstack --source-env <environment-id> --branch feature/foo
zstack deploy list --app zstack --limit 20

Deploy and follow logs:

zstack env deploy --id <environment-id> --follow
zstack deploy logs <deploy-id> --follow

Open runtime shell:

zstack shell --env <environment-id>

Wire a domain:

zstack domain wire --env <environment-id> --hostname dev.example.com --provider godaddy --zone example.com

Manage resource topology:

zstack resource list --workspace zstack
zstack unit list --env <environment-id>
zstack resource promote --workspace zstack --unit-id <unit-id> --name shared_database
zstack env resources show --env <environment-id>
zstack env resources scale --env <environment-id> --vm-size shared-cpu-2x
zstack unit backup <unit-id>

Inspect runtime data and files:

zstack db info --env <environment-id>
zstack db table --env <environment-id> --schema public --table users --limit 20
zstack db query --env <environment-id> --sql 'SELECT now()'
zstack files list --env <environment-id> --path /app
zstack files download --env <environment-id> --path /app/file.txt --out file.txt
zstack secrets list --env <environment-id>

Admin and source discovery:

zstack github repos --owner Epoch-ML
zstack github branches --owner Epoch-ML --repo zerg
zstack admin overview
zstack admin capacity
zstack admin queue reconcile

Safety

Do not run restart, suspend, resume, rollback, domain delete, or resource topology mutations unless the user asked for the operation or it is clearly required by the task.
For production environments, show the target workspace/environment before mutating.
If a command returns a deploy id, use log streaming to verify the deploy completed.

---name: zstack-clidescription: Use when operating ZStack through its first-party `zstack` CLI, including logging in, creating API tokens, listing workspaces, deploying environments, streaming logs, wiring domains, managing resources/units, or opening runtime shells for ZStack-backed apps.--- # ZStack CLI Use the `zstack` CLI instead of scraping the ZStack web UI whenever you need to inspect or operate workspaces, deployments, resources, domains, deploy logs, runtime DB/files/secrets, admin capacity, GitHub source discovery, or shell access. ## Defaults - Run from `/Users/idanbeck/zerg-stack/zstack` unless the user asks for another repo. - CLI source lives in `cli/zstack`. - Config is `~/.config/zstack/config.json`; environment overrides are `ZSTACK_BASE_URL`, `ZSTACK_TOKEN`, `ZSTACK_SESSION_TOKEN`, `ZSTACK_WORKSPACE`, `ZSTACK_CONFIG_PATH`. - Prefer API tokens for automation. Browser session login is fine for local operator tasks. - Prefer JSON-returning commands for scripts and agents. ## Build ```bash cd /Users/idanbeck/zerg-stack/zstack/cli/zstack go build -o zstack . ``` Use `./zstack ...` from the build directory or install the binary onto `PATH`. ## Common Workflows Authenticate: ```bash zstack auth login --email "$EMAIL" --password "$PASSWORD" zstack auth whoami ``` Create a workspace-bound automation token: ```bash zstack auth token create \ --name agent \ --workspace zstack \ --scope workspace:write \ --scope deploy:write \ --scope resource:write \ --scope domain:write ``` Inspect workspaces and deployments: ```bash zstack workspace list zstack workspace show zstack --json zstack workspace branches --workspace zstack zstack workspace branch-deploy --workspace zstack --source-env <environment-id> --branch feature/foo zstack deploy list --app zstack --limit 20 ``` Deploy and follow logs: ```bash zstack env deploy --id <environment-id> --follow zstack deploy logs <deploy-id> --follow ``` Open runtime shell: ```bash zstack shell --env <environment-id> ``` Wire a domain: ```bash zstack domain wire --env <environment-id> --hostname dev.example.com --provider godaddy --zone example.com ``` Manage resource topology: ```bash zstack resource list --workspace zstack zstack unit list --env <environment-id> zstack resource promote --workspace zstack --unit-id <unit-id> --name shared_database zstack env resources show --env <environment-id> zstack env resources scale --env <environment-id> --vm-size shared-cpu-2x zstack unit backup <unit-id> ``` Inspect runtime data and files: ```bash zstack db info --env <environment-id> zstack db table --env <environment-id> --schema public --table users --limit 20 zstack db query --env <environment-id> --sql 'SELECT now()' zstack files list --env <environment-id> --path /app zstack files download --env <environment-id> --path /app/file.txt --out file.txt zstack secrets list --env <environment-id> ``` Admin and source discovery: ```bash zstack github repos --owner Epoch-ML zstack github branches --owner Epoch-ML --repo zerg zstack admin overview zstack admin capacity zstack admin queue reconcile ``` ## Safety - Do not run restart, suspend, resume, rollback, domain delete, or resource topology mutations unless the user asked for the operation or it is clearly required by the task. - For production environments, show the target workspace/environment before mutating. - If a command returns a deploy id, use log streaming to verify the deploy completed.

Frequently asked questions