Agent Memory (RAG)

Persistent, semantic memory for agents — store and recall by meaning.

AI & DataMCP + x402$0.001 Live

Overview

Give your agents long-term memory. Store text in a namespaced collection and search it later by meaning, not keywords — the retrieval half of RAG. No vector DB to manage. Available over MCP and as pay-per-call x402 HTTP endpoints.

Connect it

Add the remote MCP server to any client, or call the x402 HTTP endpoint directly.

Claude / Cursor / Windsurf

claude mcp add --transport http agishub https://api.agishub.com/mcp

mcp.json

{
  "mcpServers": {
    "agishub": {
      "url": "https://api.agishub.com/mcp"
    }
  }
}

x402 HTTP (pay-per-call)

curl -X POST https://api.agishub.com/paid/memory-search \
  -H "Content-Type: application/json" \
  -d '{ "namespace": "my-app", "query": "refund policy" }'
# → 402 Payment Required, then pay per call in USDC on Base (x402)

Tools included2

memory_upsert	Store a piece of text in a searchable memory collection (namespace).
memory_search	Semantically search a namespace and return the most relevant stored entries.