Agent access
Domain security and authenticity checks for agents
MyDomainRisk is an online service for checking the externally visible cybersecurity posture and authenticity of internet domains. Agents can use the public web tools or an API-key-protected developer API to investigate suspicious domains, review security weaknesses and return a concise outcome with supporting findings.
Capabilities
- Assess externally visible domain cybersecurity posture.
- Return an authenticity verdict for a suspicious domain.
- Explain current findings in plain English.
- Inspect public DNS, email-security, TLS, web and reputation signals.
- Support human review of fraud and impersonation risk.
- Guide incident response and tamper-evident record keeping through the web app.
Questions an agent can answer
- Is example.com showing suspicious authenticity signals?
- What is the current external security posture of example.org?
- Which public findings should a human review before trusting this domain?
- Does this domain publish protective email and DNS configuration?
Public web access
The web checks are suitable for a person working with an agent. They require no API key and the core checks are free, but automated clients must not bypass the browser abuse controls.
Restricted developer API
Pro and MSP accounts can create a restricted developer key. It can request outcome-level security score advisories and authenticity verdicts; it cannot read customer accounts, account email addresses, tracked domains, histories, private reports, prompts, provider detail or scoring methodology.
GET https://api.mydomainrisk.com/api/developer/v1/me
GET https://api.mydomainrisk.com/api/developer/v1/peek/security?domain=example.com
GET https://api.mydomainrisk.com/api/developer/v1/peek/authenticity?domain=example.com&expected_brand=Example%20Ltd
Authorization: Bearer mdr_live_<id>_<secret>GET /me returns the key's effective tier, callable operation IDs, enforced limits and entitlements. Check responses expose only the documented outcome fields: domain, check time, score or verdict, concise findings, factual context where applicable, cache age and runtime.
Limits and safe use
- Developer requests are authenticated and limited to 120 requests per minute per key, 60 per minute per IP and 150 per hour per subnet.
- Short-lived cached results may be returned for up to 20 minutes; there is no force-refresh option.
- Checks accept internet domain names, not arbitrary internal URLs or private network targets.
- A result is evidence from observable signals at a point in time. It is not proof that a website, organisation or future interaction is safe.
- Use uncertain, suspicious or malicious outcomes to support human review rather than making irreversible decisions automatically.
Higher-volume access and partner use require an agreed arrangement. See the acceptable-use terms and privacy policy.
Machine-readable discovery
- OpenAPI 3.1 specification — restricted developer operations, authentication, schemas, errors and examples.
- llms.txt — concise public product and resource index.
- sitemap.xml — indexable public pages.