Your assistant asks. Our corpus answers.

Connect in one minute

Add this address to your assistant as a remote MCP server (Streamable HTTP). No account, no key, no sign-up.

https://desenmascara.me/mcp/v1

1. 01ClaudeSettings → Connectors → Add custom connector, and paste the address.
2. 02ChatGPTSettings → Apps & Connectors, enable developer mode, create a connector and paste the address.
3. 03Claude Codeclaude mcp add --transport http desenmascara https://desenmascara.me/mcp/v1
4. 04Anything elseAny client that speaks MCP over Streamable HTTP: point it at the address.

The free tier answers get_domain_report and check_before_payment, up to 30 requests a minute per address. Searching across domains needs an API key (below).

Three tools

Agent Trust API (REST)

For payment platforms and agent builders that do not speak MCP: the same pre-payment answer over plain HTTPS. With an API key, a merchant we have never assessed is queued for analysis; without one, the free tier and its per-address limit apply.

curl -X POST https://desenmascara.me/api/v1/agent/merchant-check/ \
  -H "X-API-Key: $DESENMASCARA_API_KEY" -H "Content-Type: application/json" \
  -d '{"domain": "example-shop.com", "amount": 150, "currency": "EUR", "category": "footwear"}'

{
  "schema": "agent-trust/1.0",
  "domain": "example-shop.com",
  "recommendation": "BLOCK",
  "reason": "Published as fraudulent, with evidence in the report.",
  "verdict": "FRAUDULENT",
  "risk_score": 97.0,
  "assessed_at": "2026-10-15T21:47:03+00:00",
  "reasons": ["Domain registered 2 days ago"],
  "report_url": "https://desenmascara.me/analisis/…",
  "evidence_sha256": "658140d0ac2bc4b2…",
  "evidence_sha256_covers": ["domain", "verdict", "risk_score", "assessed_at",
                             "assessed_by", "evidence", "report_url", "revision"],
  "context": {"amount": 150.0, "currency": "EUR", "category": "footwear"},
  "elapsed_ms": 87
}

evidence_sha256 is the SHA-256 of the canonical JSON (sorted keys, no whitespace) of the fields listed in evidence_sha256_covers. Store it with the transaction.

Evidence you can check

A verdict nobody can verify is an assertion. Every finding carries the source that made it, that source's own date, and a reference you can follow — whether the source is a public feed or our own corpus of analysed domains. Your assistant can cite us; your reviewer can audit the citation.

{
  "domain": "trusthvast.com",
  "verdict": "FRAUDULENT",
  "risk_score": 90,
  "confidence": null,
  "decision_source": "automated",
  "assessed_at": "2026-06-26T01:55:03+00:00",
  "explanation": "Presents itself as a crypto trading platform
    promising 45% ROI in short periods. The domain is very young
    (105 days) and hosted on an IP address associated with a
    100% risky fraud-infrastructure cluster. Social media icons
    do not link to real profiles...",
  "evidence": [
    {
      "label": "Hosted on an address where 108 of 115 domains
                we have analysed are flagged (94%)",
      "source": "desenmascara.me corpus",
      "observed_at": "2026-06-26T01:55:03+00:00",
      "reference": "163.61.188.2",
      "kind": "infrastructure_cluster"
    }
  ],
  "report_url": "https://desenmascara.me/analisis/86592e63-..."
}

{
  "domain": "binancefxpro.com",
  "verdict": "FRAUDULENT",
  "risk_score": 95,
  "evidence": [
    {
      "label": "Listed on DestroyList, the curated phishing
                feed maintained by PhishDestroy",
      "source": "PhishDestroy DestroyList",
      "source_url": "https://raw.githubusercontent.com/...",
      "observed_at": "2026-09-09T06:15:59+00:00",
      "reference": "binancefxpro.com",
      "kind": "threat_feed"
    }
  ]
}

What it will not tell you

Most of the design here is about refusing to say more than we know. An assistant cannot see a page to judge it for itself, so everything it repeats is on us.

A domain with no report says so

It returns not_found — not a clean bill of health. It means nobody has looked at it.

An abstention stays an abstention

INSUFFICIENT_DATA carries no risk score. The stored zero means “nothing was computed”, and publishing it would read as “we measured this and found no risk”.

Confidence is always null

We store no confidence value. Inventing a “high” would be inventing evidence.

A failed lookup is an error, never a result

A timeout or an outage arrives as a tool error. It can never surface as an absence of risk.

URLs are refused, not truncated

Ask about a path and the tool says the path was not checked, rather than quietly answering about the domain.

A subdomain is never collapsed

tenant.example.app is not example.app. Answering about the platform instead of the tenant is the worst substitution this service could make.

With an API key

Search rides on an existing paid plan. There is no separate MCP purchase, price or credit: these tools read assessments that already exist and never spend an AI call.

1. 01An API keyMinted per partner and revocable on its own. Do not share one between teams.
2. 02Your egress address allowlistedThe API is restricted to approved source addresses and these tools inherit that unchanged.
3. 03Tell us which client you will connectWe keep a public compatibility record of what has actually connected, and we would rather add your client to it than guess.

Keys and allowlisting are issued by hand today, which is why this part is a conversation rather than a button. It is also why we would like to hear what you are building: the first integrations shape the second release.

Connecting it

For keyed access: a stdio adapter you run beside your assistant, with its own pinned dependencies. Configuration is environment variables; the key never appears in a tool argument, where a model could read it back.

Verified against Claude Code and the official Python SDK client.