scalattice
OfficialClick on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@scalatticecheck my credits"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Scalattice CLI
Sign in from a terminal, land in an interactive CLI, mint an account management key (slt_mgmt_…) for automation, create an inference API key (slt_…) for OpenAI-compatible SDKs, check credits, and manage a provider fleet. Optional MCP mode for AI coding agents.
Product: scalattice.com/cli · npm: scalattice-cli · Cloud docs: developers#cli · providers#fleet-api · Installer: scalattice.cloud/install/cli
Install
# Paste-friendly
curl -fsSL https://scalattice.cloud/install/cli | sh
# Or with npm (Node 18+)
npm install -g scalattice-cliOr sign in on Cloud /auth and run the curl command shown there. That stores a session in ~/.config/scalattice/config.json, signs the browser tab in, installs this CLI if needed, and opens the prompt.
Related MCP server: Modal MCP Server
Quick start
scalatticeYou get a scalattice> prompt. Type commands without the scalattice prefix:
scalattice> whoami
scalattice> login
scalattice> credits
scalattice> help
scalattice> exitlogin stores a session only. For the OpenAI SDK, create a key and export it:
scalattice developers keys create
eval "$(scalattice init)" # after OPENAI_API_KEY is in the envOne-shot from any shell still works: scalattice login, scalattice whoami, and the rest.
Then use any OpenAI SDK:
from openai import OpenAI
client = OpenAI() # uses OPENAI_BASE_URL + OPENAI_API_KEY
print(client.models.list())Keys
Prefix | Role |
| Account management: credits, inference-key CRUD, fleet |
| Inference only (chat/completions via |
| Machine agent token ( |
Create management keys on Account or account keys create if you need a long-lived secret for MCP. The CLI itself uses your session.
Commands
Inside the prompt, drop the scalattice prefix. From a normal terminal, keep it.
Command | What it does |
| Interactive prompt ( |
| Session only (what |
`developers keys list | create |
`account keys list | create |
| Print OpenAI env exports (needs |
| Wallet + model grants |
| Show session / account |
| List fleet (id, status, token last four) |
| Add a machine (prints |
| New token for an existing machine (prints once) |
| Remove a never-connected machine |
| Fleet earnings |
| Pause or resume all machines |
| Patch one machine’s schedule |
| MCP stdio server (run as |
developer/developers, provider/providers, machine/machines, and key/keys are aliases.
Config: ~/.config/scalattice/config.json (mode 0600) — session + email only. Keys are never written there.
Env overrides: SCALATTICE_CLOUD_URL, SCALATTICE_API_URL, SCALATTICE_API_KEY, SCALATTICE_MGMT_KEY, SCALATTICE_SESSION_TOKEN.
Provider fleet
scalattice login
scalattice provider machines create
scalattice provider machines
scalattice provider pause # or resumeSee Fleet API docs.
MCP (optional)
MCP is not a second install. After login, leave the prompt and run scalattice mcp. For a headless agent without a session, set SCALATTICE_MGMT_KEY in that environment.
{
"mcpServers": {
"scalattice": {
"command": "scalattice",
"args": ["mcp"]
}
}
}License
MIT
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP-first toolbox for agents: KV storage, auth, queue, and utility tools. Free in early access.
Shared memory and actions for Claude, Kiro, OpenAI, Cursor, and other MCP-compatible AI clients.
- QuallaaOAuthcom.quallaa
Talk to your public-facing AI from any MCP client — Claude, ChatGPT, Cursor, Cline, Windsurf.
OpenAI-compatible LLM MCP (7 tools); chat via balance key or x402 USDC on Base
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceUnified CLI multiplexer for AI coding agents, enabling orchestration of multiple coding CLI tools through a single interface with session persistence, cost tracking, and MCP tool support.13 npm19MIT
- FlicenseBqualityDmaintenanceConnects AI agents to Modal.com infrastructure, allowing management of deployments, apps, containers, volumes, secrets, and environments via MCP tools.30-
- AlicenseNot gradedqualityCmaintenanceLocal-first harness and ticket operations for AI assistants via CLI or MCP.2 npmMIT
- AlicenseBqualityAmaintenanceProvides data science, plotting, statistics, system, and web tools via MCP, with interactive CLI clients for OpenAI-compatible and Anthropic providers.261MIT