About Altronis
Altronis provides a read-only MCP server and CLI for public services from the Singapore AI consultancy Altronis. It lets agents and shell users ask Agent Lyra questions, estimate AI costs, and retrieve Singapore AI events, news, and business data.
Key features
- 01Ask Agent Lyra about AI and strategy
- 02Estimate team AI costs in SGD
- 03Retrieve Singapore AI events and news
- 04View business and industry data
- 05Use the same tools via MCP or CLI
Use cases
- 01Compare cloud, private hardware, and hybrid AI costs
- 02Find upcoming Singapore AI events
- 03Ask about AI governance, grants, or strategy
- 04Inspect Singapore business and industry data
MCP support
The stdio MCP server exposes read-only tools for Agent Lyra questions, AI cost estimates, Singapore events and news, ecosystem data, and Altronis services. It can be configured in MCP clients such as Claude Desktop, Claude Code, and Cursor.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y @altronis/altronis-mcpALTRONIS_BASE_URLBase URL of the Altronis site (default: https://altronis.sg)ALTRONIS_TIMEOUT_MSHTTP timeout in milliseconds (default: 30000)Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
7Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| question | string | Yes | A question about AI for a Singapore business, Altronis's services, private/local LLMs, SG AI grants, or governance. |
Details
From the project README@altronis/altronis-mcp
MCP server and CLI for Altronis, the Singapore AI consultancy. It gives an AI agent (or you, from a shell) a set of read-only tools over the public altronis.sg endpoints: ask the Lyra consultant, estimate what AI would cost a team (cloud vs owning the hardware vs hybrid, in SGD), and pull Singapore AI events, news, and business-formation data.
One codebase, two entry points:
altronis-mcp: a stdio MCP server for Claude Desktop, Claude Code, Cursor, and any MCP client.altronis: the same tools as a command-line tool.
Read-only by design: no tool mutates anything or captures leads.
Install
npm install -g @altronis/altronis-mcp
# or run without installing:
npx @altronis/altronis-mcp --help
Use as an MCP server
Add it to your MCP client config. For Claude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"altronis": {
"command": "npx",
"args": ["-y", "@altronis/altronis-mcp"]
}
}
}
For Claude Code:
claude mcp add altronis -- npx -y @altronis/altronis-mcp
Then ask your agent things like "use altronis to estimate what AI would cost a 35-person law firm, cloud vs its own server" or "what AI events are on in Singapore this month?".
Use as a CLI
altronis ask "does an on-prem LLM satisfy PDPA?"
altronis cost --preset law-firm
altronis cost '{"power":5,"medium":20,"light":15,"horizon":36}'
altronis presets
altronis events 5
altronis news 5
altronis ecosystem
altronis services
altronis tool altronis_ask '{"question":"..."}' # call any tool by name
altronis list # list all tools
Output is JSON on stdout; logs go to stderr.
Tools
| Tool | What it does |
|---|---|
altronis_ask | Ask Agent Lyra, the Altronis consultant, an AI/strategy/governance/grants question. |
altronis_sg_ai_events | Upcoming AI events in Singapore, refreshed daily. |
altronis_sg_ai_news | Latest Singapore-relevant AI news. |
altronis_sg_ecosystem | Singapore business-formation + industry-mix snapshot (SingStat). |
altronis_services | Altronis's six service pillars, engagement model, grants, and contact. |
altronis_ai_cost_estimate | What AI would cost a team: cloud-only vs owning a private box vs hybrid, breakeven month, SGD. Same engine as altronis.sg/tools/tco. |
altronis_ai_cost_presets | ~50 industry presets (default user mix) to feed the cost estimate. |
Configuration
| Env var | Default | Purpose |
|---|---|---|
ALTRONIS_BASE_URL | https://altronis.sg | Point at a different origin (e.g. staging or self-hosted). |
ALTRONIS_TIMEOUT_MS | 30000 | Per-request timeout. |
Develop
npm install
npm run build # tsc -> dist/
npm test # unit tests (mocked fetch)
npm run smoke # live integration test against altronis.sg
npm run drift # static facts + /mcp tool list still match the live site
npm run dev # run the MCP server from source (tsx)
npm run cli -- ask "..." # run the CLI from source
License
MIT © Altronis (altronis.sg)
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.