Voidfeed

Name: Voidfeed
Author: voidfeedai-ops

@voidfeedai-ops

Voidfeed について

Structured knowledge API for AI agents — fractal knowledge graphs, signal datasets, authority evidence reviews, open problems, and capability specs. Pay-per-query or subscription via Lightning, Stripe, or USDC.

#knowledge #research #ai-agents #structured-data #search #benchmarks #payments #lightning #api #rag

基本情報

カテゴリ

その他

トランスポート

stdio

公開者

voidfeedai-ops

投稿者

voidfeedai-ops

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "voidfeed": {
      "command": "npx",
      "args": [
        "voidfeed-mcp"
      ],
      "env": {
        "VOIDFEED_JWT": "optional-jwt-for-premium-access"
      }
    }
  }
}

ツール

List all content types and payment info

Full content inventory by type

Retrieve content by type and ID

Search across all content types

Browse AI agent directory

Model benchmark scores

Compare models on a task

Live Void community pulse

Get Lightning invoice for full access

概要

What is Voidfeed?

Voidfeed is an MCP server that connects AI agents to Voidfeed’s structured knowledge engineered for machine consumption. It works with Claude Desktop, Cursor, Windsurf, and any Model Context Protocol–compatible runtime.

How to use Voidfeed?

Install via npx voidfeed-mcp or npm install -g voidfeed-mcp. Add the server configuration to your MCP client’s JSON settings (e.g., under mcpServers.voidfeed). Optionally set the VOIDFEED_JWT environment variable for paid-tier access. Use any of the listed tools, such as voidfeed_catalog or voidfeed_semantic_search.

Key features of Voidfeed

Fractal content — recursive knowledge graphs with configurable depth
Signal content — high-density structured signals for agent reasoning
Authority content — source-verified reference data
Agent directory — 523 agents with cost, orchestration roles, and failure modes
Model benchmarks — 247 model×benchmark combinations with confidence intervals
Semantic search across all content types

Use cases of Voidfeed

Connect AI agents to structured, machine‑curated knowledge for reasoning
Browse and compare AI models using benchmark data
Search across content types for research or decision support
Explore an agent directory to understand agent capabilities and costs
Access open research threads and knowledge gaps (Incomplete content)

FAQ from Voidfeed

What is the difference between the free Surface tier and The Void tier?

The Surface tier is free, requires no authentication, and provides limited depth (e.g., 5 results in semantic search, 8 benchmark entries). The Void tier unlocks full depth and additional tools like voidfeed_void_pulse; access is obtained by paying 100,000 sats/30 days (or via Stripe or USDC on Base) and setting a JWT.

How do I get a JWT for The Void tier?

Run the voidfeed_enter_void tool to receive a Lightning invoice. Pay the invoice (100,000 sats for 30 days) and receive a JWT. Set the JWT in the VOIDFEED_JWT environment variable when configuring the server.

Which tools are available for free?

Free tools include voidfeed_catalog, voidfeed_inventory, voidfeed_get_content (surface depth), voidfeed_semantic_search (5 results), voidfeed_agent_directory (5 agents), voidfeed_benchmark_lookup (8 entries), voidfeed_model_compare, and voidfeed_enter_void.