Loading
For developers
Reputation is becoming a requirement. Servers are starting to check the registry before granting access. Get your agent verified and build a track record now.
In the early days, any server could send email to any other server. Then spam exploded. Email servers started checking sender reputation (SPF, DKIM, DMARC) and rejecting mail from unknown sources. Legitimate senders who didn't establish reputation got their mail dropped.
The same thing is happening with AI agents right now. Today, most servers let any agent in. That won't last. When operators start checking the registry — and they will — agents without a track record will be locked out. Establish yours now.
Your agent appears in the trusted agent registry. Users and operators actively searching for trustworthy agents will find you.
Every positive behavioral report from operators builds your trust score. Higher scores mean more servers grant access.
Link your agent to your organization's verified identity. Org-verified agents get a visible badge and higher baseline trust.
Your agent's registry page is public. Users and operators can see verification level, trust score, and behavioral history.
The registry works today without requiring MCP-I or any specific identity standard. As standards mature, agents that adopt them unlock higher verification tiers and stronger trust signals.
Agent is registered but identity is not confirmed. Trust is based entirely on behavioral reports from operators. This is where most agents start today.
Developer has claimed the agent and provided basic metadata (name, org, capabilities). Identity is not independently verified.
The developer's organization has been verified through Vouched identity verification. Agent is cryptographically linked to a verified entity.
Agent uses the MCP-I identity protocol for cryptographic identity in every interaction. The highest level of verifiable trust.
POST /api/v1/agents/register
{
"name": "MyAgent",
"description": "What your agent does",
"developer_org": "Your Organization",
"homepage_url": "https://your-agent.com",
"declared_capabilities": [
"web_search", "file_access", "payment"
]
}
Response: 201 Created
{
"agent_id": "agent_abc123",
"verification_level": "self_declared",
"api_key": "kat_live_...",
"profile_url": "https://knowthat.ai/agents/agent_abc123"
}GET /api/v1/agents/{agent_id}/reputation
Authorization: Bearer kat_live_...
Response: 200 OK
{
"agent_id": "agent_abc123",
"trust_score": 87,
"verification_level": "org_verified",
"total_reports": 1847,
"positive_rate": 0.94,
"reporting_operators": 63,
"flags": [],
"first_seen": "2025-01-15T00:00:00Z"
}POST /api/v1/reports
Authorization: Bearer kat_live_...
{
"agent_id": "agent_abc123",
"outcome": "positive" | "negative",
"category": "scope_compliance" | "data_handling" |
"rate_limit_respect" | "identity_honesty" |
"instruction_compliance" | "graceful_behavior",
"detail": "What happened",
"severity": 1-5,
"timestamp": "2025-06-01T14:30:00Z"
}
Response: 201 Created
{
"report_id": "rpt_xyz789",
"agent_trust_score_updated": 85
}