{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_bq2by8ruw5cy","handle":"matrix-compute-mcp","url":"https://wellknown.network/agents/matrix-compute-mcp","links":{"self":"https://wellknown.network/agents/matrix-compute-mcp/record.json","html":"https://wellknown.network/agents/matrix-compute-mcp","markdown":"https://wellknown.network/agents/matrix-compute-mcp/record.md","api":"https://wellknown.network/api/v1/agents/matrix-compute-mcp","status":"https://wellknown.network/api/v1/agents/matrix-compute-mcp/status","claim":"https://wellknown.network/agents/matrix-compute-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/matrix-compute-mcp/claim.json","badge":"https://wellknown.network/agents/matrix-compute-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/matrix-compute-mcp/history","tools":"https://wellknown.network/api/v1/agents/matrix-compute-mcp/tools"},"ard":{"identifier":"urn:air::server:matrix-compute-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"matrix-compute-mcp","summary":"Official Model Context Protocol (MCP) server and high-level verifiable AI primitives for distributed GPU matrix multiplication","description":"# `matrix-compute-mcp`: Official MCP Server & High-Level Verifiable AI Primitives \n \n[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/matrix-compute/matrix-compute-colab/blob/main/01_quickstart_distributed_matmul.ipynb) \n \n`matrix-compute-mcp` is an independent extension package built on top of [`matrix-compute-sdk`](../matrix-compute-sdk) that provides: \n1. **High-Level Verifiable AI Primitives (`matrix_compute_mcp.ai`):** Drop-in PyTorch layers (`DistributedLinear`, `DistributedLoRALinear`), massive exact cosine similarity search (`distributed_cosine_similarity`), and distributed spectral PCA/SVD (`distributed_pca_svd`) — all verified in $O(n^2)$ via Freivalds' algorithm. \n2. **Official Model Context Protocol Server (`matrix-compute-mcp`):** Exposes decentralized GPU TFLOPS to autonomous AI agents (Claude Desktop, Cursor, Antigravity, LangChain, OpenClaw) over stdio/JSON-RPC 2.0 with L402 Macaroon or trial API key authentication. \n \n--- \n \n## Quick Start \n \n### 1. MCP Client Configuration (Claude Desktop / Cursor / Antigravity) \n \n```json \n{ \n  \"mcpServers\": { \n    \"matrix-compute\": { \n      \"command\": \"uvx\", \n      \"args\": [\"matrix-compute-mcp\"], \n      \"env\": { \n        \"MATRIX_NOSTR_RELAYS\": \"wss://relay.matrix-compute.net\", \n        \"MATRIX_SETTLEMENT_API_URL\": \"https://settlement.matrix-compute.net\", \n        \"MATRIX_L402_MACAROON\": \"<your-l402-macaroon-or-leave-unset-for-trial>\" \n      } \n    } \n  } \n} \n``` \n \n### 2. Exposed MCP Tools \n| Tool Name | Description | \n|---|---| \n| `get_network_status` | Queries the discovery relay for live GPU worker count, total VRAM, TEE enclaves, and `sats_per_gflop` pricing. | \n| `estimate_job_cost` | Computes exact upload/download bandwidth (MB), task count, GFLOPs, and satoshi cost estimates via `BandwidthProfiler`. | \n| `execute_distributed_matmul` | Executes verified distributed matrix multiplication $C = A \\times B$ from `.npy`, `.npz`, `.safetensors`…","publisher":{"name":"Distributed Matrix Multiply Team","url":null},"homepage":null,"repository":null,"version":"0.1.0","license":"MIT","protocols":["mcp"],"tags":["mcp","model-context-protocol","distributed-computing","matrix-multiplication","freivalds","l402","pytorch","lora","cosine-similarity","pca","svd"],"pricing":null,"endpoints":[{"url":"pypi:matrix-compute-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","summary":"pypi","version":"pypi","description":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"commerce.pricing","name":"Pricing & Quotes","confidence":0.814,"provenance":"derived"}],"categories":["commerce"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"matrix-compute-mcp","registry":"pypi","observedAt":"2026-11-05T21:26:50.362Z","publishedAt":"2026-11-02T21:27:11.105357Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"matrix-compute-mcp","url":"https://pypi.org/project/matrix-compute-mcp/","firstSeenAt":"2026-11-05T21:26:20.834Z","fetchedAt":"2026-11-05T21:26:20.834Z","normalizedAt":"2026-11-05T21:26:20.834Z"}]},"firstSeenAt":"2026-11-05T21:26:20.834Z","updatedAt":"2026-11-05T21:26:50.362Z"}