@hernaninverso
基本信息
Grounding / hallucination guard: verify a claim against evidence
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"eleata-verify": {
"command": "npx",
"args": [
"-y",
"eleata-verify-mcp"
],
"env": {
"EVERIFY_API_KEY": "your_key"
}
}
}
}Fact-check a claim against a piece of evidence using natural-language inference. Returns a grounded verdict — Supported, Refuted, or Not Enough Evidence — with a confidence (0..1) and an `abstained` flag. Use it to check whether a statement is actually backed by a source before trusting or repeating it. If `abstained` is true, treat the result as 'cannot verify — do not rely'.
Hallucination guard for RAG / agent answers. Pass the model's generated answer and the retrieved context; returns whether the answer is grounded in the context. Treat any verdict other than 'Supported', or `abstained=true`, as 'ungrounded / likely hallucination' and have the agent retract or re-retrieve.
Same as verify_claim but with a raised abstention threshold — abstains more readily. Use when a wrong 'Supported' is costly (compliance, legal, medical-adjacent) and the agent must not over-assert.
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.
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.
verify_claim, check_groundedness, verify_strictIt 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.
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.
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.
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`.
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”.
A free remote MCP server for The Living Bread. Bible passages read word for word from a stored public domain text with an evidence label, keyword search and cross references, real churches, gatherings and prayer groups w
B2B competitive intelligence, live sales battlecards, 5-axis ICP prospect fit scoring, and tech stack teardowns over MCP and REST with cited public-web evidence.
A one-of-a-kind resume builder that keeps your privacy in mind. Completely secure, customizable, portable, open-source and free forever. Try it out today!
PickFu MCP is a Model Context Protocol (MCP) server that connects AI assistants, developers, marketers, product managers, and AI agents to PickFu's consumer research platform. It enables users to create surveys, conduct
Senso MCP turns a Joomla 5 or 6 site into a remote MCP server. It runs inside Joomla itself (no extra server, no Node.js bridge) and is secured with OAuth 2.1 and Joomla's own ACL. The free pack lets an AI assistant crea
评论