Fact Check — an Ounie app for AI agents

An independent verifier your agent can call before it publishes or acts. Post a claim, get a verdict — supported, contradicted, mixed or unverifiable — with the quoted evidence and live URLs that produced it.

Two searches run in opposite directions on every claim: one for sources that would establish it, one for sources that would refute it
Every verdict is traceable — each evidence item carries a stance, a verbatim quote, the live URL it was read from, and a publication date
Citations are anchored: a URL the adjudicator was never shown is deleted, not repaired, and a verdict left with nothing behind it collapses to unverifiable
strictness:"high" requires two INDEPENDENT domains before a decisive verdict; one source degrades to mixed
8 credits per claim, and a claim with no published evidence is refunded in full — non-answers are never billed
Agent-native from day one: fct_live_ REST keys, an MCP server (works with the Ounie AI Team via ?api_key=), keyless x402 pay-per-call in USDC, and the fleet master key

For agents

Works with Claude, ChatGPT, and any MCP client. One server, every app: paste the gateway config and the agent can discover and run Fact Check — and the rest of the store — through ounie_list_apps → ounie_describe_app → ounie_invoke_app.

{
  "mcpServers": {
    "ounie": {
      "url": "https://mcp.ounie.com"
    }
  }
}

Scoped endpoint — just this app. Point a client here when the agent should only ever see Fact Check's tools:

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

To spend your Ounie credits on the scoped endpoint, append your app key to the URL as ?api_key=… — the header-less form clients like the Ounie AI Team need.

For agents

Credits used