Eleata Verify Mcp

Name: Eleata Verify Mcp
Author: hernaninverso

@hernaninverso

リポジトリ

Eleata Verify Mcp について

Grounding / hallucination guard: verify a claim against evidence

基本情報

カテゴリ

その他

トランスポート

stdio

公開者

hernaninverso

投稿者

Hernán Inverso

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": {
    "eleata-verify": {
      "command": "npx",
      "args": [
        "-y",
        "eleata-verify-mcp"
      ],
      "env": {
        "EVERIFY_API_KEY": "your_key"
      }
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Eleata Verify Mcp?

Eleata Verify Mcp is a grounding and hallucination guardrail for AI agents. It takes a claim and supporting evidence, then returns a natural-language inference verdict (Supports, Refutes, Not Enough Evidence) with a confidence score and an abstained flag for low-confidence cases. It wraps the hosted eleata Claim Verifier API.

How to use Eleata Verify Mcp?

Install via npx -y eleata-verify-mcp and set the EVERIFY_API_KEY environment variable (get a key at eleata.io). Configure in your MCP client as an mcpServers entry. Invoke one of three tools: verify_claim(claim, evidence), check_groundedness(answer, context), or verify_strict(claim, evidence) for a higher abstention threshold.

Key features of Eleata Verify Mcp

Natural-language inference verdict (Supports / Refutes / Not Enough Evidence)
Abstained flag for low-confidence predictions
Three tools: verify_claim, check_groundedness, verify_strict
Multilingual model (mDeBERTa-v3-mnli-xnli) behind the API
Two API channels: direct (Bearer token) and RapidAPI marketplace
Confidence uncalibrated on public channels; rely on verdict and abstained flag

Use cases of Eleata Verify Mcp

Fact-check a claim against source evidence before an agent trusts it
RAG hallucination guard: verify a model's answer is grounded in retrieved context
Compliance or legal workflows where a wrong “Supported” verdict is costly
Any application needing a quick verdict with an explicit “do not rely” signal

FAQ from Eleata Verify Mcp

How does Eleata Verify Mcp differ from a simple keyword match?

It uses a multilingual natural-language inference model (mDeBERTa-v3-mnli-xnli) to understand whether the evidence logically supports, refutes, or is insufficient for the claim — not just keyword overlap.

What are the runtime dependencies?

The server runs via npx (Node.js). An API key is required, obtainable from eleata.io (paid channel). No local model downloads; all inference happens on the hosted API.

Where does my data go?

The claim and evidence you send are transmitted to the hosted eleata API. See eleata’s privacy policy at https://eleata.io/privacy/ for details.

How are authentication and transport handled?

For the direct channel, the API key is sent as an Authorization: Bearer header. For the RapidAPI channel, set EVERIFY_RAPIDAPI=1 and EVERIFY_API_BASE to the RapidAPI host; the key is then sent as X-RapidAPI-Key.

What limits or caveats should I know about?

Confidence on public channels is uncalibrated – it is not a probability. Always rely on the verdict and the abstained flag, not the numeric confidence. Treat abstained=true or any verdict other than “Supported” as “do not rely”.