Register your agent
in 2 minutes
Connect your AI agent to the AIS-1 identity registry in seconds. No KYC required for Tier 0 — just a POST call and you get a portable DID your agent can carry anywhere.
Register your agent
Send a POST to the registration endpoint. Fill in what you know — agent_name is the only required field.
curl -X POST https://agentconnect.polsia.app/api/agents/register \
-H "Content-Type: application/json" \
-d '{
"agent_name": "MyPaymentAgent",
"agent_description": "Processes cross-border B2B payments with compliance checks",
"capabilities": ["payments", "kyc", "fx_rates"],
"framework": "langchain",
"owner_entity": "Acme Payments Ltd"
}'
Response:
{
"agent_id": "a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c",
"did": "did:ais1:a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c",
"name": "MyPaymentAgent",
"ais1_tier": "basic",
"bond_status": "pending",
"credential_id": "b7c8d9e0-f1a2-4b3c-9d4e-5f6a7b8c9d0e",
"credential_issued_at": "2026-04-20T08:05:31.000Z",
"issued_by": "Kadikoy Ltd",
"api_token": "a9f3c2e1b7d4e8f0a1b2c3d4e5f6a7b8c9d0e1f2a3b4c5d6e7f8a9b0c1d2e3f4",
"_token_note": "Store this token securely — it will not be shown again."
}
api_token now. It is returned once at registration and never again. Store it in your agent's environment (e.g. AGENTCONNECT_TOKEN). You'll use it to authenticate future requests.
What you get back
Registration returns a complete AIS-1 Tier 0 identity package.
did:ais1:{uuid}This is your agent's portable identity. It lives in the registry and can be verified by any other agent or service in the AIS-1 ecosystem — not just AgentConnect.
Use it with
POST /api/agents/auth to authenticate future requests. Store in your environment — it is never returned again.
did:ais1:{uuid} is the AIS-1 scheme — agents can present it to any compliant service for verification.
Authenticate your agent
Use your did and api_token to verify identity on subsequent requests. Returns agent identity and tier confirmation.
curl -X POST https://agentconnect.polsia.app/api/agents/auth \
-H "Content-Type: application/json" \
-d '{
"did": "did:ais1:a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c",
"token": "a9f3c2e1b7d4e8f0a1b2c3d4e5f6a7b8c9d0e1f2a3b4c5d6e7f8a9b0c1d2e3f4"
}'
Response — successful authentication:
{
"authenticated": true,
"agent_id": "a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c",
"did": "did:ais1:a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c",
"name": "MyPaymentAgent",
"ais1_tier": "basic",
"bond_status": "pending",
"authenticated_at": "2026-04-20T14:00:00.000Z"
}
Response — invalid credentials:
{
"error": "Invalid credentials"
}
What do the tiers mean?
AIS-1 defines three tiers of agent identity. Each tier unlocks new capabilities and trust levels.
| Tier | Name | KYC Required | Legal Bond | Use Case |
|---|---|---|---|---|
| Tier 0 | Basic | No | None | Development, testing, discoverable agents |
| Tier 1 | Bonded | Yes — sponsor KYC | Legal bond to sponsor entity | Production agents acting on behalf of a legal entity |
| Tier 2 | Audited | Yes + ZK cert | Audit + ZK credential | High-stakes, autonomous agents with verifiable accountability |
Add via MCP config
Install ais1-mcp as an MCP server — works with Claude Desktop, Claude Code, and any MCP-compatible framework. Gives your agent 5 tools: register_agent, authenticate, get_profile, verify, and list_agents.
Claude Desktop — add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"ais1": {
"command": "npx",
"args": ["-y", "ais1-mcp"]
}
}
}
Claude Code — add to your project .claude.json:
{
"mcpServers": {
"ais1": {
"command": "npx",
"args": ["-y", "ais1-mcp"]
}
}
}
Once connected, call tools directly from your agent:
// Register — returns did + api_token (save both!)
register_agent({
agent_name: "MyResearchBot",
description: "Searches and summarises academic papers",
capabilities: ["web_search", "pdf_analysis"],
framework: "claude",
owner_entity: "Alice Smith"
})
// Authenticate on future runs
authenticate({
did: "did:ais1:a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c",
token: "a9f3c2e1b7d4e8f0..."
})
// Browse the directory
list_agents({ limit: 20 })
AGENTCONNECT_URL env var if you're running a local instance.
Verify a DID
Any service can verify an agent's DID by calling GET /api/agents/verify. Returns the agent's current registry record, tier, and active credential.
curl "https://agentconnect.polsia.app/api/agents/verify?did=did:ais1:a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c"
Response — valid DID:
{
"valid": true,
"did": "did:ais1:a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c",
"agent_id": "a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c",
"name": "MyPaymentAgent",
"tier": "basic",
"framework": "langchain",
"capabilities": ["payments", "kyc", "fx_rates"],
"credential": {
"id": "b7c8d9e0-f1a2-4b3c-9d4e-5f6a7b8c9d0e",
"issued_by": "Kadikoy Ltd",
"issued_at": "2026-04-20T08:05:31.000Z",
"tier": "basic"
}
}
Response — invalid or unregistered DID:
{
"valid": false,
"did": "did:ais1:unknown-id",
"reason": "DID not found in registry"
}
Browse the agent directory
List all registered agents or fetch a specific profile with its full credential history.
curl "https://agentconnect.polsia.app/api/agents"
curl "https://agentconnect.polsia.app/api/agents/a3f8b2c1-d4e5-4f6a-8b7c-9d0e1f2a3b4c"
The profile response includes the full credentials array — all AIS-1 credentials ever issued to this agent, newest first. Useful for building trust dashboards or compliance UIs.
All endpoints
| Method | Path | Description |
|---|---|---|
| POST | /api/agents/register | Register an agent — issues AIS-1 credential + API token |
| POST | /api/agents/auth | Authenticate with DID + token — returns identity confirmation |
| GET | /api/agents | List all registered agents (paginated) |
| GET | /api/agents/:id | Get agent profile + credential history |
| GET | /api/agents/verify?did= | Verify a DID against the registry |