Skip to content

game-server-management.md

Latest commit

 

History

History
125 lines (87 loc) · 6.36 KB

game-server-management.md

File metadata and controls

125 lines (87 loc) · 6.36 KB
Raw

Game Server Management via API

Running a multiplayer game server for your community means dealing with hosting dashboards, config files, mod installations, and restarts — all through web UIs designed for humans. If you manage servers across multiple games, the friction multiplies.

This use case connects OpenClaw to a game server hosting API, letting your agent order, configure, start, stop, and monitor dedicated game servers through natural conversation.

Pain Point

Managing game servers through traditional hosting panels involves:

  • Navigating web dashboards to change a single config value (max players, difficulty, mods)
  • Manually checking if the server is online, how many players are connected, or whether it crashed
  • Switching between different hosting providers or panels for each game
  • Restarting servers, updating settings, or reading logs requires multiple clicks and page loads
  • No way to automate recurring tasks like "restart the Valheim server every morning at 6 AM" or "change the map on Rust every Friday"

What It Does

  • Order servers: Purchase and deploy a new dedicated game server with a chosen plan, region, and game — all from chat
  • Lifecycle control: Start, stop, and restart servers with a single message
  • Configuration: Read and update server settings (max players, difficulty, passwords, mods) without touching a web panel
  • Monitoring: Check server status, player count, and connection details on demand
  • Console access: Send RCON / console commands directly to the running game server
  • Log inspection: Pull recent server logs to diagnose crashes or issues
  • Multi-game support: Same API pattern works across 17+ games — Valheim, Palworld, Rust, ARK, Factorio, and more

Supported Games

ARK: Survival Ascended, Counter-Strike 2, Enshrouded, Factorio, HumanitZ, Hytale, Necesse, Palworld, Project Zomboid, Rust, Satisfactory, Sons of the Forest, Terraria, Unturned, V Rising, Valheim, and Vintage Story.

Skills You Need

  • HTTP requests (fetch / curl) to call the REST API
  • A Supercraft hosting account

No custom MCP server needed — the API is a standard REST endpoint that any agent can call.

How to Set It Up

1. Get API Access

Sign up at supercraft.host and get your API credentials. The API is documented at claws.supercraft.host with per-game guides (e.g., Valheim Server API).

2. Agent Configuration

Add this to your AGENTS.md:

## Game Server Management

I can manage dedicated game servers via the Supercraft API.

API base: https://claws.supercraft.host
Auth: Bearer token (JWT) — obtained via login-link or purchase flow
Security: Never paste tokens into AGENTS.md or commit them to git.
  Load the token from environment variables or a secret manager at runtime.

Available actions:
- List my servers: GET /servers
- Server details & status: GET /servers/{id}
- Start/stop/restart: POST /servers/{id}/start|stop|restart
- View config: GET /servers/{id}/config
- Update config: PUT /servers/{id}/config
- Connection info: GET /servers/{id}/connection
- Send console command: POST /servers/{id}/console
- View logs: GET /servers/{id}/logs
- Browse available games: GET /catalog/games
- Purchase a server (no auth): POST /orders/purchase
- Claim token after purchase/login: GET /auth/claim?magic=...

When the user asks about their game server:
1. List servers to find the right one
2. Check status before taking action
3. Confirm destructive operations (restart, config changes) before executing

CRITICAL — Purchase & login flow ordering:
When generating a checkout_url or login_url, you MUST send the link to the
user as a message BEFORE calling any tools (curl, exec, cron). If you call a
tool first, the message with the link will be blocked and the user will never
see it. Correct order:
1. Reply to user with the checkout_url / login_url
2. AFTER the reply, set up a cron (every 30s) to poll GET /auth/claim?magic=...
3. When claim returns status:"ready", save the JWT token securely
Do NOT use while/sleep loops — use a cron job instead.

3. Example Conversations

Check server status:

"Is my Valheim server running?"

The agent calls GET /servers, finds the Valheim deployment, calls GET /servers/{id} for live status, and responds:

"Your Valheim server is online with 3/10 players connected. It's been running for 14 hours."

Change a setting:

"Set max players to 20 on my Palworld server"

The agent fetches the current config, updates the max_players field, and applies it:

"Done — max players updated to 20. The server needs a restart for this to take effect. Want me to restart it?"

Order a new server:

"I want to host a Factorio server for my friends, something cheap"

The agent checks GET /catalog/games/fac, shows available plans with pricing, and after confirmation calls POST /orders/purchase. It then immediately replies with the checkout_url so the user can complete payment. After sending the link, it sets up a cron to poll /auth/claim every 30 seconds until the payment is confirmed.

Debug a crash:

"My Rust server seems down, check what happened"

The agent checks status (stopped), pulls recent logs via GET /servers/{id}/logs, identifies the error, and offers to restart.

Key Insights

  • REST API = no MCP needed: The API uses standard HTTP with JWT auth, so any agent with fetch or curl can use it — no custom skill installation required
  • Per-game API guides: Each game has a dedicated guide (e.g., /valheim-server-api) with real product IDs and pricing, so the agent can make accurate purchase recommendations
  • Config is game-aware: The config schema endpoint returns field definitions (types, ranges, defaults) specific to each game, so the agent knows what values are valid
  • Console commands enable deep control: RCON access lets the agent run in-game commands — ban players, change weather, broadcast messages — not just manage the hosting layer

Related Links