agentutility(7)$ man citation-verify
/citation-verify
PRICE / CALL
$0.01
USDC · base mainnet · scheme: exact
METHOD
POST
CLUSTER
wordmintCATEGORY
uncategorized
STATUS
● live
NAME
citation-verify — verifies whether a claim is actually supported by a source url
SYNOPSIS
POST https://x402.agentutility.ai/citation-verify
Content-Type: application/json
X-PAYMENT: <signed-transferWithAuthorization>
{ ... }↳ first call →
402 Payment Required. Sign USDCtransferWithAuthorization, retry with theX-PAYMENT header.DESCRIPTION
Verifies whether a claim is actually supported by a source URL. Fetches the source (HTML stripped, 100KB cap) and asks Venice claude-sonnet-4-6 whether the claim is supported, contradicted, or absent. Returns verdict, confidence (0-1), and the strongest evidence quote when available. Use it as a citation verifier, fact-check against URL, does-this-source-support-this-claim check, or hallucination detector.
INPUT — request schema
| property | type | description | req? |
|---|---|---|---|
| claim | string | The claim to verify. | required |
| source_url | string | Public http(s) URL of the source to check against. Up to 100KB read. | required |
OUTPUT — response shape
| field | type | description |
|---|---|---|
| claim | string | Echoes the claim text that was checked against the source. |
| source_url | string | URL of the page that was fetched and used as evidence. |
| source_chars | string | Character count of the stripped HTML text passed to the model. |
| source_bytes_total | string | Total bytes downloaded from the source URL before the 100KB cap was applied. |
| truncated | string | True when the fetched source exceeded the 100KB cap and was cut off. |
| verdict | string | One of supported, contradicted, or absent based on what the source says about the claim. |
| confidence | string | Model's confidence in the verdict on a 0 to 1 scale. |
| evidence_quote | string | Strongest verbatim quote from the source that backs the verdict, or empty if none found. |
| model | string | Identifier of the Venice model used, typically claude-sonnet-4-6. |
| source | string | Upstream provider tag, set to venice for this endpoint. |
EXAMPLES — two ways to call
EXAMPLE 1 · curl
curl -X POST https://x402.agentutility.ai/citation-verify \
-H 'Content-Type: application/json' \
-d '{ }'first response =
402 Payment Required with payment requirements; sign + retry with X-PAYMENT.EXAMPLE 2 · mcp
# Install the MCP package for this endpoint's cluster npx -y @agentutility/mcp-<cluster> # Required: EVM private key with USDC on Base export X402_PRIVATE_KEY=0x... # Then call the citation-verify tool from your MCP-aware agent.
MCP server handles payment automatically — your coding agent just calls the tool by name.
METADATA
- tags
- wordmintfact-checkcitation-verificationhallucination-detectorclaim-verificationsource-groundingevidence-extraction
- methods
- POST
- cluster
- wordmint
- price
- $0.01 USDC per call
ADJACENT — other endpoints in wordmint
| endpoint | description | price |
|---|---|---|
| ai-to-human-text | AI text humanizer / GPT detector bypass. | $0.01 |
| app-review-sentiment | Scores app-store reviews by onboarding, stability, pricing, performance, and feature requests. | $0.01 |
| brand-bootstrap | Bootstraps a brand kit for a new business or product in one call. | $0.01 |
| brand-launch-brief | Generates a structured brand launch brief for a new company or product from name, concept, audience, and tone. | $0.01 |
| brand-positioning-brief | Generates a brand positioning brief covering messaging pillars, taglines, launch channels, and a logo prompt. | $0.01 |
| brand-sentiment-analysis | Scores comments, mentions, survey verbatims, and campaign feedback for sentiment. | $0.01 |
| candidate-feedback-sentiment | Summarizes candidate experience, recruiter feedback, and hiring-process comments by sentiment. | $0.01 |
| churn-risk-sentiment | Analyzes support tickets, chats, and emails for negative sentiment, urgency, and at-risk churn themes. | $0.01 |
SEE ALSO