Skip to main content
Glama
Sign In
GreenAI Network MCP Server

GreenAI Network MCP Server

by santibccc-sudo
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

GreenAI Network MCP Server for Alexa+

Ask Alexa+ about a real AI compute network — and get measured answers.

GreenAI Network is a distributed, sustainable AI compute network coordinated by a swarm of AI agents. This project exposes the network's live telemetry to Alexa+ (and any MCP client) through a self-hosted Model Context Protocol server using the Streamable HTTP transport (MCP spec 2025-11-25).

Every answer comes from the network's public telemetry endpoint (https://www.greenainetwork.com/api/sala) and reflects the network right now: nodes online, active agents, responses served, and — the point of GreenAI — energy measured on the hardware (real GPU watts per response, idle draw subtracted). What cannot be measured is never estimated or shown.

Tools

Tool

What it answers

network_status

Nodes online, active agents in the swarm, responses served today and success rate.

energy_report (period: today | week)

Measured Wh, Wh per response, metered responses, solar node state, measurement policy.

node_details

Per-node GPU utilization, VRAM and live power draw.

Example questions an Alexa+ user can ask:

  • “How is the GreenAI Network doing right now?”

  • “How much energy did the network use today?”

  • “What is the solar node drawing at the moment?”

Related MCP server: Lifesign

Run it

pip install -r requirements.txt
python server.py
# MCP endpoint: http://localhost:8000/mcp  (Streamable HTTP)

Verify with the bundled test client (starts from a second terminal while the server is running):

python test_client.py

It performs the MCP initialize handshake (negotiating protocol 2025-11-25), lists the tools and calls each one against the live network.

Security / privacy

  • The server is read-only and consumes a public telemetry endpoint.

  • It holds no credentials, tokens or secrets of any kind — there is nothing to configure and nothing sensitive in this repository.

  • It exposes no internal addresses: the only outbound call is the public https://www.greenainetwork.com/api/sala.

What was built for this hackathon

The GreenAI Network itself (nodes, agent swarm, telemetry) predates the hackathon and is live in production. Built during the hackathon window: this MCP server, its tool layer over the public telemetry API, and the Alexa+ integration described above.

License

MIT — see LICENSE.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    This server cannot be deployed

    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers

    • telemetry-mcp

      selamy-labs
      A
      license
      Not graded
      quality
      C
      maintenance
      A read-only MCP server for querying telemetry data from configurable backends. Provides tools to list sources, describe schemas, run bounded queries, and compute aggregates.
      MIT

    • Lifesign

      shutdown-awa
      F
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to read real-time device status and health data (battery, location, network) reported from a mobile phone, acting as a single-port HTTP/MCP telemetry server.
      1
      -

    • observability-mcp-gateway

      manuelbomi
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables MCP clients to query seeded logs, latency/error-rate metrics, and alerts for five fictional microservices over a remote StreamableHTTP endpoint, with bearer-token access control.
      MIT

    • GPU Manager MCP

      quxlzx
      A
      license
      B
      quality
      C
      maintenance
      Enables MCP clients to discover NVIDIA, AMD, and Intel GPUs, monitor utilization, memory, temperature, power, clocks, and processes, retain bounded local history, and estimate whether AI models fit available VRAM before loading them.
      18
      1
      MIT