sir
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., "@sirask GPT-4 for a second opinion on this plan"
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.
$SIR plugin for Claude Code
Frontier models below list price, billed to your $SIR balance, from inside Claude Code.
claude plugin marketplace add mc1d/sir-plugin
claude plugin install sir@sir
export SIR_BASE_URL=https://<your gateway>/v1
export SIR_API_KEY=sk-sir-...
claudeWhat you get
| Every model on the gateway, its lab and list price |
| Ask any model: second opinions from GPT, Gemini or DeepSeek, cheap bulk work on |
| What the key can spend, and 30 days of spend by model |
| Your referral link and commission |
| Config for Cursor, Codex, Cline, Aider, opencode, Claude Desktop and any OpenAI SDK |
| Make a key by signing locally with a Solana keypair. No sign-up |
| Point Claude Code itself at $SIR, so every request is billed there |
| Balance at a glance |
| Have another lab's model review the current plan or diff |
Related MCP server: ask-another
Run Claude Code on $SIR
/sir:setup writes this for you, or set it yourself:
export ANTHROPIC_BASE_URL=https://<your gateway> # no /v1: Claude Code adds /v1/messages
export ANTHROPIC_AUTH_TOKEN=$SIR_API_KEY
export ANTHROPIC_API_KEY=
claudeClaude model names map to $SIR models: opus to sir-frontier-plus, sonnet to sir-frontier, haiku to sir-base.
Other tools
The same MCP server runs anywhere MCP does:
{ "mcpServers": { "sir": { "command": "npx", "args": ["-y", "github:mc1d/sir-plugin"], "env": { "SIR_API_KEY": "sk-sir-...", "SIR_BASE_URL": "https://<your gateway>/v1" } } } }sir/server/sir-mcp.mjs is a single bundled file with no dependencies, built from the $SIR monorepo. Needs Node 20 or newer.
This server cannot be deployed
Maintenance
Related MCP Connectors
Generate text, images, speech, music, and video with any AI model, from one credit balance.
Discover, run, inspect, build, test, and privately reuse AI workflows.
Account, billing, team, API key and model discovery tools for agents that use ModelsLab.
101Image, video, music and text generation across 100+ models through one endpoint.
Related MCP Servers
AlicenseAqualityAmaintenanceOne workspace for every AI coding assistant. Governance tools for API lint, diff, persistent ledger, multi-model deliberation, security audit, and test verification. Works with Claude Code, Codex, Cursor, and Gemini CLI.42134,323 npm24MIT- AlicenseAqualityBmaintenanceEnables AI assistants to query hundreds of models across OpenAI, Google, and OpenRouter, with model discovery, deep-research jobs, and image generation.103MIT
- -licenseNot gradedqualityNot gradedmaintenanceEnables discovery of public AI models with pricing, documentation context, and OpenAI-compatible integration examples. Supports both read-only queries and paid async media generation tasks.-
- AlicenseNot gradedqualityBmaintenanceEnables chatting with AI models, listing and filtering models, retrieving model details and credit balance, and checking generation costs and provider statistics.159 npmMIT