SKILL.md
Jack Basics
This skill teaches you how to work effectively in Jack Cloud projects.
What This Skill Provides
- How to query and modify Jack databases (D1)
- How to deploy and check status
- How to use Jack MCP tools when available
- Common patterns and troubleshooting
Key Concept
Jack projects use cloud infrastructure. Databases run on Cloudflare's edge network, not locally. Use Jack tools for all operations.
Database Operations
Query Data
jack db execute "SELECT * FROM users LIMIT 10"Modify Data
jack db execute --write "INSERT INTO users (name, email) VALUES ('Alice', '[email protected]')"View Schema
jack db execute "SELECT name FROM sqlite_master WHERE type='table'"
jack db execute "PRAGMA table_info(users)"Create Table
jack db execute --write "CREATE TABLE posts (id INTEGER PRIMARY KEY, title TEXT, created_at TEXT DEFAULT CURRENT_TIMESTAMP)"MCP Tools (Preferred When Available)
Check if mcpjack* tools are in your available tools:
| Tool | Purpose |
|---|---|
mcpjackexecute_sql |
Query or modify database |
mcpjacklist_databases |
List project databases |
mcpjackcreate_database |
Create new D1 database |
Example MCP usage:
mcp__jack__execute_sql
sql: "SELECT * FROM users"
allow_write: falseFor writes, set allow_write: true.
Deployment
Deploy to Production
jack shipCheck Deployment Status
jack infoShows: live URL, last deploy time, attached services.
Stream Production Logs
jack logsLocal Development
jack devRuns locally with remote bindings (real D1, KV, R2).
MCP Tools
| Tool | Purpose |
|---|---|
mcpjackdeploy_project |
Deploy to production |
mcpjackgetprojectstatus |
Check deployment status |
mcpjacktail_logs |
Stream production logs |
Services
List All Services
jack servicesCreate Services
jack services db create # D1 database
jack services kv create # KV namespace
jack services storage create # R2 bucket
jack services vectorize create # Vector indexMCP Tools
| Tool | Purpose |
|---|---|
mcpjackcreate_database |
Create D1 database |
mcpjackcreatestoragebucket |
Create R2 bucket |
mcpjackcreatevectorizeindex |
Create vector index |
Secrets
Set a Secret
jack secrets set STRIPE_SECRET_KEYPrompts for value (hidden input).
Set Multiple Secrets
jack secrets set API_KEY WEBHOOK_SECRETList Secrets
jack secrets listProject Structure
Typical Jack project:
my-project/
├── src/
│ └── index.ts # Worker entry point
├── wrangler.jsonc # Cloudflare config (bindings, routes)
├── package.json
└── .jack/
└── project.json # Jack project linkwrangler.jsonc- Defines D1 bindings, KV namespaces, environment.jack/project.json- Links to Jack Cloud (managed) or your Cloudflare account (BYO)
Troubleshooting
"Database not found"
Create one:
jack services db create"Not a Jack project"
Link the directory:
jack link --byoMCP tools not available
Install MCP config:
jack mcp installThen restart your AI assistant.
Need to see what went wrong
Stream production logs:
jack logsQuick Reference
| Task | Command |
|---|---|
| Query database | jack db execute "SELECT ..." |
| Write to database | jack db execute --write "INSERT ..." |
| Deploy | jack ship |
| Check status | jack info |
| Stream logs | jack logs |
| Local dev | jack dev |
| Create database | jack services db create |
| Set secret | jack secrets set KEY |
Summary
- Use
jackcommands for all operations - Use MCP tools when available (
mcpjack*) - Databases are remote - D1 runs on Cloudflare's edge
- Deploy with
jack ship- handles build and deployment - Check logs with
jack logs- debug production issues