Are you an LLM? Read llms.txt for a summary of the docs, or llms-full.txt for the full context.
Skip to content

OpenClaw / Claude Code

Deploy from your agent. Jack Cloud gives your AI agent everything it needs to create APIs, add databases, and ship to production — no config, no dashboard.

Install the Skill

npx clawhub@latest install jack-cloud

This drops the skill into .claude/skills/jack-cloud/ and your agent picks it up automatically next conversation.

Browse on ClawHub: clawhub.ai/hellno/jack-cloud

What Your Agent Learns

One skill. Full cloud deployment.

  • Create projects — jack new my-api scaffolds, configures, and deploys in one command
  • Ship changes — jack ship builds and pushes to production in seconds
  • Add a database — jack services db create provisions SQLite and wires the bindings
  • Query data — jack services db execute "SELECT * FROM users" runs against production
  • Stream logs — jack logs for real-time debugging
  • Manage secrets — jack secrets set API_KEY without touching a dashboard

Plus storage, vector search, cron jobs, and custom domains when you need them.

Works with MCP tools or CLI. No Cloudflare account required.

See also: AI agents guide for MCP setup and full tool reference.

Prerequisites

Your agent needs jack installed:

curl -fsSL docs.getjack.org/install.sh | bash

Or try without installing: npx -y @getjack/jack new my-api --template api

Authentication

Already have a token? Set it and go:

export JACK_API_TOKEN=jkt_abc123...

No token yet? Use magic auth (no browser needed):

jack login --email [email protected]

Check your email for a 6-digit code, then:

jack login --email [email protected] --code 847291

The CLI is now authenticated and a jkt_* API token is created automatically. See the AI agents guide for the full breakdown.

CLI or MCP — Both Work

The skill teaches CLI-first commands, with MCP tool equivalents noted throughout. If your agent has mcp__jack__* tools available (via jack mcp serve), it will prefer those automatically.

TaskCLIMCP Tool
Deployjack shipmcp__jack__deploy_project
Statusjack infomcp__jack__get_project_status
Query DBjack services db execute "..."mcp__jack__execute_sql
Create DBjack services db createmcp__jack__create_database
Logsjack logsmcp__jack__tail_logs

JSON Output for Agents

Key commands support --json for machine-readable output:

jack ship --json
# {"success":true,"projectName":"my-api","url":"https://my-api.runjack.xyz","deployMode":"managed"}
 
jack services db execute --json "SELECT count(*) as n FROM users"
# {"success":true,"results":[{"n":42}],"meta":{"duration_ms":12}}

Example: Agent Deploys an API

Here's what a typical agent workflow looks like after installing the skill:

Agent: I'll create a new API project with a database.
 
> jack new my-api --template api
> jack services db create
> jack services db execute --write "CREATE TABLE items (id INTEGER PRIMARY KEY, name TEXT, created_at TEXT DEFAULT CURRENT_TIMESTAMP)"
 
Now I'll add a route to query items and deploy.
 
> [edits src/index.ts]
> jack ship
 
Your API is live at https://my-api.runjack.xyz/api/items

Source

The skill is open source: github.com/getjack-org/skills