mcpField
P
pgiq-mcp

pgiq-mcp

Local
github.com/powergridiqBusiness & Marketing0Updated 4w ago
Repository

About pgiq-mcp

pgiq-mcp connects MCP-capable agents to PowerGridIQ’s read-only power-market API. It provides cited ratings and grid data to help teams assess locations and operating times for large electricity loads.

Key features

  1. 01List rated markets
  2. 02Five-pillar market ratings with evidence
  3. 03Rank markets by decision lens
  4. 04Recent grid developments
  5. 05Grid snapshots and price windows

Use cases

  1. 01Compare markets for a large electricity load
  2. 02Review a market’s rating and supporting evidence
  3. 03Find low-cost hours for a flexible load

MCP support

The local stdio MCP server exposes tools backed by the PowerGridIQ read-only REST API, including market ratings, developments, grid snapshots, and price windows.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "powergridiq-pgiq": {
      "command": "npx",
      "args": [
        "-y",
        "pgiq-mcp"
      ],
      "env": {
        "PGIQ_KEY": "<PGIQ_KEY>",
        "PGIQ_API": "<PGIQ_API>"
      }
    }
  }
}
Run commandnpx -y pgiq-mcp
Environment variables to set
PGIQ_KEYSecretOptional PowerGridIQ API key (pk_live_...) for a higher quota.
PGIQ_APIOptional API base override.

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmpgiq-mcp1.3.0

Tools

11

Saved tool list · updated 2w ago

This tool takes no parameters.

Details

From the project README

PowerGridIQ MCP server

Exposes the PGIQ Rating as tools any MCP-capable agent (Claude Desktop, Claude Code, Cursor, custom agents) can call: "where should I site or schedule a large electricity load, and how is a given market rated." It wraps the read-only REST API at https://powergridiq.com/api/v1, covering 67 global power markets with a reasoned, cited decision, not just raw feeds.

Tools

  • pgiq_markets - list every rated market (id, tier, score, outlook). Start here.
  • pgiq_rating(market) - full five-pillar rating + evidence for one market (e.g. quebec).
  • pgiq_best(lens?, group?, min_tier?, limit?) - the ranked best-market decision. lens is default|cost|carbon|momentum.
  • pgiq_developments(market?) - recent cited grid developments.
  • pgiq_grid(market) - grid snapshot: carbon intensity, price, fuel mix, demand, firm headroom, peak stress. Merges live signals where available, else a clearly labelled modelled baseline.
  • pgiq_cheapest_window(market?) - cheapest hours to run a flexible load (modelled daily price shape, local time). Omit market for the cheapest grids ranked.

Install & run

cd pgiq-mcp
npm install
node index.js   # speaks MCP over stdio

Configure in Claude Desktop

Add to claude_desktop_config.json (Settings → Developer → Edit Config):

{
  "mcpServers": {
    "powergridiq": {
      "command": "node",
      "args": ["/absolute/path/to/pgiq-mcp/index.js"],
      "env": { "PGIQ_KEY": "pk_live_optional_key_for_higher_limits" }
    }
  }
}

Restart Claude Desktop, then ask things like "Use PowerGridIQ to find the best carbon-light market for a 100 MW AI cluster in Europe" or "What's the PGIQ rating for Ireland and why?"

Authentication & limits

  • No key needed for the free tier: 100 requests/day per IP.
  • For a higher monthly quota, set PGIQ_KEY=pk_live_... (developer 25,000/mo, pro 500,000/mo). Request a key at [email protected].
  • Override the API base with PGIQ_API if needed.

Notes

  • The rating is a directional screen, not a connection guarantee; the interconnection queue is the real gate. Every market carries a confidence level, treat lower confidence with more caution.
  • Live grid signals, rating-change webhooks, and metered keys are available now; short-horizon forecasts are next.
  • A hosted remote MCP server (no local install) is planned; for now this runs locally via stdio.

Links

Comments

Markdown supported

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.