Terros MCP Server
Click 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., "@Terros MCP Server list the first 10 users with an admin role"
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.
@terros-inc/mcp
Production-ready MCP server scaffold for the Terros platform. This initial version implements User API tools.
Implemented tools
terros_get_current_user- Get the authenticated user profile (GET /users/me)terros_list_users- List users with optionalrole,limit, andoffsetterros_get_user- Get a user by ID
Requirements
Node.js 20+
npm
Terros API key
Setup
pnpm install
cp .env.example .env
pnpm build
pnpm startEnvironment variables
Variable | Required | Default | Description |
| Yes | - | API key sent as |
| No |
| Base URL for Terros API |
MCP client configuration
Claude Desktop
Add this to Claude Desktop MCP config:
{
"mcpServers": {
"terros": {
"command": "node",
"args": ["/absolute/path/to/terros-mcp/dist/index.js"],
"env": {
"TERROS_API_KEY": "<your_key>",
"TERROS_API_BASE_URL": "https://api.terros.com/v1"
}
}
}
}Cursor
Configure MCP server entry:
{
"name": "terros",
"command": "node",
"args": ["/absolute/path/to/terros-mcp/dist/index.js"],
"env": {
"TERROS_API_KEY": "<your_key>",
"TERROS_API_BASE_URL": "https://api.terros.com/v1"
}
}OpenClaw
Example server launch:
TERROS_API_KEY=<your_key> TERROS_API_BASE_URL=https://api.terros.com/v1 node dist/index.jsDevelopment
pnpm devBuild
pnpm buildAvailable Tools
6 toolsterros_get_companyA
Get a company from Terros by company ID. Omit companyId to get the current user's company.
Inputs: companyId.
terros_get_current_userB
Get the authenticated user's own profile from Terros.
Inputs: no named parameters.
terros_get_teamB
Get a Terros team by ID, including child teams and members.
Inputs: teamId (required).
terros_get_userB
Get a Terros user by User ID (e.g. U:abc123), email address, or clientUserId.
Inputs: userId (required).
terros_list_teamsC
List teams in the Terros workspace. Filter by company or archived status.
Inputs: companyId, showArchived.
terros_list_usersB
List users in the Terros workspace. Filter by company, team, specific user IDs, or archived status.
Inputs: companyId, teamId, userIds, showArchived.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
Every tool has a clearly distinct purpose targeting specific resources: get_company, get_current_user, get_team, get_user, list_teams, and list_users. There is no overlap in functionality, as each tool handles a unique entity or action, making it easy for an agent to select the correct one without confusion.
All tool names follow a consistent 'terros_verb_noun' pattern, using snake_case throughout. The verbs (get, list) are used predictably, with 'get' for retrieving single entities and 'list' for collections, ensuring a uniform and readable naming convention across the toolset.
With 6 tools, the server is well-scoped for managing users, teams, and companies in a workspace. Each tool earns its place by covering essential operations without bloat, providing a focused and manageable interface for the domain, which is typical for such administrative functions.
The toolset covers read operations (get and list) for users, teams, and companies, but lacks create, update, or delete functionalities. This is a notable gap, as agents cannot perform full lifecycle management, though they can work around it by using existing data for queries and reporting.
Related MCP Connectors
Users, organizations, roles, permissions, applications and feature flags.
Manage a ScaledThought account: API keys, budgets, usage and prepaid top-ups. Needs an API key.
Manage API keys, identities, permissions, rate-limit overrides and verification analytics.
Account, billing, team, API key and model discovery tools for agents that use ModelsLab.