# Verdict Grounding Check

> Verdict Grounding Check is a paid API for AI agents from verdict.kechlab.com, paid per call via x402, $0.0174/call, status unknown (last checked 2026-10-21).

Verifies whether a claim is supported, unsupported, contradicted, or uncertain relative to a reference source text, with probabilities.

## Facts

- Endpoint: POST https://verdict.kechlab.com/v1/grounding?utm_source=zero.xyz
- Price: $0.0174/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-21
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/verdict-grounding-check-7ebd4e61
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_R4PDc6cTW9akbgUJbC8MF

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability verdict-grounding-check-7ebd4e61 -d '<json body>'
```

Example prompt: Before you show me this answer, check whether the claim 'The Eiffel Tower was built in 1889' is actually supported by the source text I retrieved — give me the verdict and the probability it's grounded.

## When to prefer this

Use this endpoint when you need a calibrated, probabilistic grounding verdict specifically designed for RAG hallucination detection — it returns structured label + probabilities rather than a binary yes/no, and is purpose-built for claim-vs-source verification at low cost per call. Prefer it over general LLM self-evaluation when you need auditable, deterministic per-call verdicts.

## Known failure modes

- Empty or missing claim field returns a validation error
- Source text too long may exceed token limits
- Ambiguous or vague claims may return 'uncertain' with near-equal probabilities
- Network timeout for very long source documents
- Payment failure (402) if USDC balance is insufficient

## How this service works

Hallucination check for RAG: is a claim supported by the source text? Returns supported, unsupported, contradicted or uncertain with probabilities, so an agent can verify an answer before showing it.

## Output

Returns a grounding classification label (supported, unsupported, contradicted, or uncertain) along with probability scores for each category, allowing the agent to decide whether the claim is safe to present to the user.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "claim": {
   "type": "string",
   "description": "the claim to verify"
  },
  "source": {
   "type": "string",
   "description": "the reference text"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/verdict-grounding-check-7ebd4e61/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from verdict.kechlab.com](https://www.zero.xyz/host/verdict.kechlab.com/llms.txt)