# Agent Evidence API

> Agent Evidence API is a paid API for AI agents from agent-evidence.thx93.dev, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-22).

Verifies claims and fact-checks statements against up to 5 user-supplied public web URLs, returning cited evidence passages with support/contradiction/inconclusive assessments.

## Facts

- Endpoint: POST https://agent-evidence.thx93.dev/mcp?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-22
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agent-evidence-api-5ebab3ee
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_bCqF1Oyu0dy3XT_fIHGz-

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 agent-evidence-api-5ebab3ee -d '<json body>'
```

Example prompt: Can you fact-check the claim that 'Acme Corp manufactures centrifugal pumps' against these two pages: https://acmecorp.com/about and https://acmecorp.com/products — and tell me whether each page supports, contradicts, or can't settle it, with the exact quote used?

## When to prefer this

Choose this endpoint when you need verifiable, cited evidence from specific URLs rather than a broad web search. It is ideal when you already have candidate source URLs and need to confirm or refute a precise claim with quoted excerpts and provenance metadata (URL, timestamp, content hash). Prefer it over general search APIs when auditability and citation integrity matter — for journalism, compliance, legal research, or agent pipelines that must show their work. The per-call pricing and free sample tool make it low-risk to evaluate before committing.

## Known failure modes

- URL is inaccessible or returns non-200 status — source listed with error status, no evidence excerpts
- Claim is too vague to match any passage — assessment returns 'unsettled' with empty excerpts
- Paywall or JavaScript-rendered page blocks content retrieval — source returns empty evidence
- No sources provided or all URLs fail — empty sources array, no charge applied
- Network timeout on a URL — that source omitted or flagged with retrieval error

## How this service works

Verify a claim or fact check a statement against public web sources: send a question and up to 5 URLs, get cited evidence - passages that support, contradict or fail to settle it. Claim verification with a citation for every excerpt: source URL, retrieval time, content hash. Never charges when nothing is retrieved. Free to evaluate: the `sample` tool returns one real response at no cost, so you can judge the output before paying for your own question.

## Output

Returns a list of sources, each with their HTTP status, retrieved timestamp, content hash, and one or more evidence excerpts tagged as 'direct', or other relevance levels. A top-level assessment object summarizes whether the claim is supported, contradicted, or unsettled, with a plain-language basis statement. A limitations array notes caveats such as deterministic matching vs. semantic reasoning. Nothing is returned (and no charge applies) when no content is retrieved.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "type",
    "toolName",
    "inputSchema"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "mcp"
    },
    "method": {
     "type": "string"
    },
    "example": {
     "type": "object"
    },
    "toolName": {
     "type": "string"
    },
    "transport": {
     "enum": [
      "streamable-http"
     ],
     "type": "string"
    },
    "description": {
     "type": "string"
    },
    "inputSchema": {
     "type": "object"
    }
   },
   "additionalProperties": false
  },
  "output": {
   "type": "object",
   "required": [
    "type"
   ],
   "properties": {
    "type": {
     "type": "string"
    },
    "example": {
     "type": "object"
    }
   }
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "sources": [
   {
    "title": "Company X - About",
    "status": 200,
    "evidence": [
     {
      "excerpt": "Company X manufactures centrifugal pumps at its facility.",
      "relevance": "direct"
     }
    ],
    "final_url": "https://company.example/about",
    "retrieved_at": "2026-01-01T00:00:00.000Z",
    "content_hash_sha256": "9f2c…"
   }
  ],
  "assessment": {
   "basis": "1 source matched the question.",
   "status": "supported"
  },
  "limitations": [
   "Assessment uses deterministic lexical matching, not semantic reasoning."
  ]
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agent-evidence-api-5ebab3ee/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agent-evidence.thx93.dev](https://www.zero.xyz/host/agent-evidence.thx93.dev/llms.txt)