KS
Killer-Skills

f3-dev-servers — Categories.community

v1.0.0
GitHub

About this Skill

Perfect for Full-Stack Agents needing simultaneous development server management for Caddy, API, and Map services. F3 Nation Monorepo

F3-Nation F3-Nation
[0]
[0]
Updated: 2/27/2026

Quality Score

Top 5%
49
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add F3-Nation/f3-nation

Agent Capability Analysis

The f3-dev-servers MCP Server by F3-Nation is an open-source Categories.community integration for Claude and other AI agents, enabling seamless task automation and capability expansion.

Ideal Agent Persona

Perfect for Full-Stack Agents needing simultaneous development server management for Caddy, API, and Map services.

Core Value

Empowers agents to run Caddy, API, and Map development servers in parallel using Shell commands, leveraging `block_until_ms: 0` for immediate backgrounding and `working_directory` for repo root setup, while utilizing protocols like pnpm for package management.

Capabilities Granted for f3-dev-servers MCP Server

Starting Caddy servers for local development
Debugging API services with `pnpm -F f3-nation-api dev`
Launching Map development servers with `PORT=3000` configuration

! Prerequisites & Limits

  • Requires Shell command execution capabilities
  • Dependent on pnpm package manager
  • Limited to local development environment
Project
SKILL.md
3.7 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8

# Tags

[No tags]
SKILL.md
Readonly

F3 Nation Development Servers

Start the three dev servers needed for local f3-nation development.

Starting All Services (Agent Method)

The agent CAN start all three services by running Shell commands with block_until_ms: 0 to background them immediately.

Run these in parallel with block_until_ms: 0 and working_directory set to the repo root:

bash
1# 1. Caddy 2caddy run --config Caddyfile 3 4# 2. API 5PORT=3001 pnpm -F f3-nation-api dev 6 7# 3. Map 8PORT=3000 pnpm -F f3-nation-map dev

Note: nvm/rbenv/jenv should be auto-loaded via ~/.zshenv so no prefix needed.

Agent workflow:

  1. Run check-status.sh first to see what's already running
  2. Start only the services that aren't running (use block_until_ms: 0 for each)
  3. Wait 5 seconds for services to initialize
  4. Run check-status.sh again to verify all services are up
  5. Optionally read terminal output files to check for startup errors

Quick Check: Are Services Running?

Run the status check script from the repo root:

bash
1.cursor/skills/f3-dev-servers/scripts/check-status.sh

This script uses nc (netcat) for fast port checking (~1 second). It checks ports 443, 3000, and 3001.

Or check manually:

bash
1# Fast check with netcat 2nc -z localhost 443 && echo "Caddy OK" || echo "Caddy DOWN" 3nc -z localhost 3000 && echo "Map OK" || echo "Map DOWN" 4nc -z localhost 3001 && echo "API OK" || echo "API DOWN"

Note: Avoid using lsof for port checks on macOS - it takes 30+ seconds.

Manual Startup (User Method)

If the user prefers to start manually in named terminals:

1. Caddy (reverse proxy) - in caddy terminal

bash
1caddy run --config Caddyfile

Caddy proxies:

  • https://map.f3nation.testlocalhost:3000
  • https://api.f3nation.testlocalhost:3001

2. API Server - in api terminal

bash
1PORT=3001 pnpm -F f3-nation-api dev

3. Map App - in map terminal

bash
1PORT=3000 pnpm -F f3-nation-map dev

Startup Order

Start in this order for best results:

  1. Caddy first (handles HTTPS)
  2. API second (the map app depends on API endpoints)
  3. Map app last

Verifying Startup Success

After starting, check terminal outputs for these success indicators:

  • Caddy: Look for "msg":"serving initial configuration"
  • API: Look for ✓ Ready in X.Xs
  • Map: Look for ✓ Ready in X.Xs

Stopping Services

To stop services, use pkill (faster than lsof):

bash
1pkill -f "caddy run" 2pkill -f "f3-nation-api dev" 3pkill -f "f3-nation-map dev"

Or kill by port (slower, lsof takes 30+ seconds on macOS):

bash
1kill -9 $(lsof -t -i :3000) # Map 2kill -9 $(lsof -t -i :3001) # API 3sudo kill -9 $(lsof -t -i :443) # Caddy (needs sudo for port 443)

For user-started services in IDE terminals, press Ctrl+C in the terminal.

Troubleshooting

Port already in use

bash
1# Find what's using the port 2lsof -i :3000 3 4# Kill it 5kill -9 <PID>

Node version warning

If you see this warning:

WARN  Unsupported engine: wanted: {"node":">=20.19.0"} (current: {"node":"v18.20.4"...})

The ~/.zshenv file may be missing or misconfigured. It should contain:

bash
1export NVM_DIR="$HOME/.nvm" 2[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh"

This ensures nvm is loaded for all zsh shells, including non-interactive ones from Cursor.

Caddy certificate issues

Ensure the certificate files exist in the project root:

  • map.f3nation.test+1-cert.pem
  • map.f3nation.test+1-key.pem

If missing, regenerate with mkcert:

bash
1mkcert map.f3nation.test api.f3nation.test

Database not running

If you see database connection errors, ensure PostgreSQL/Docker is running:

bash
1docker ps # Check if postgres container is running

Related Skills

Looking for an alternative to f3-dev-servers or building a Categories.community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication