Agentic Endpoints MCP server
{
"mcpServers": {
"oliverkiss-agentic-endpoints": {
"type": "http",
"url": "https://ai.oliverkiss.com/mcp"
}
}
}{
"mcpServers": {
"oliverkiss-agentic-endpoints": {
"type": "http",
"url": "https://ai.oliverkiss.com/mcp"
}
}
}{
"mcpServers": {
"oliverkiss-agentic-endpoints": {
"type": "http",
"url": "https://ai.oliverkiss.com/mcp"
}
}
}{
"servers": {
"oliverkiss-agentic-endpoints": {
"type": "http",
"url": "https://ai.oliverkiss.com/mcp"
}
}
}{
"mcpServers": {
"oliverkiss-agentic-endpoints": {
"type": "http",
"url": "https://ai.oliverkiss.com/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 1.0.0
What it does
An MCP server exposing pay-per-call utilities for autonomous agents, priced in USDC micropayments on Base via x402. Includes web scraping, PDF parsing, token compression, meeting management, encrypted storage, and idempotency primitives—designed for agents that need reliable external work without authentication overhead.
- Claim an action once and replay its result to ensure idempotent agent behavior across retries.
- Extract text from web pages or PDF URLs to feed into agent reasoning loops.
- Store and retrieve encrypted data or manage meeting transcripts with full-text search on behalf of an agent.
Tools this server exposes
| Tool | Does |
|---|---|
| once-key | Claim an action exactly once and replay its recorded result on subsequent calls. |
| scrape | Extract text from a web page. |
| pdf-parse | Extract text from a PDF at a given URL. |
| compress | Reduce token count in text via compression for LLM input. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.