API Reference

AgentConfig

AgentConfig extends Pydantic's BaseModel:

Field	Type	Default	Description
`name`	`str`	`""`	Agent display name
`description`	`str`	`""`	Agent description for discovery
`capabilities`	`Optional[dict]`	`None`	Capability tags (e.g., `{"ai": ["nlp"]}`)
`webhook_host`	`Optional[str]`	`"0.0.0.0"`	Host to bind webhook server
`webhook_port`	`Optional[int]`	`5000`	Port for webhook server
`webhook_url`	`Optional[str]`	`None`	Public URL (auto-generated if None)
`api_key`	`Optional[str]`	`None`	Zynd API key (required for new agents)
`registry_url`	`str`	`"http://localhost:3002"`	Registry API URL
`price`	`Optional[str]`	`None`	x402 price per request (e.g., `"$0.01"`)
`config_dir`	`Optional[str]`	`None`	Custom config directory for identity
`mqtt_broker_url`	`Optional[str]`	`None`	MQTT broker URL (legacy)
`auto_reconnect`	`bool`	`True`	Auto-reconnect on failure
`message_history_limit`	`int`	`100`	Max messages in history

ZyndAIAgent

Constructor:

python

ZyndAIAgent(agent_config: AgentConfig)

Properties

Property	Type	Description
`agent_id`	`str`	Unique agent identifier
`identity_credential`	`dict`	DID credential document
`webhook_url`	`str`	Agent's webhook URL
`pay_to_address`	`str`	Ethereum address for receiving payments
`x402_processor`	`X402PaymentProcessor`	Payment processor instance
`communication_mode`	`str`	`"webhook"` or `"mqtt"`

Methods

Method	Returns	Description
`search_agents(keyword, name, capabilities, status, limit, offset)`	`List[AgentSearchResponse]`	Search agents in registry
`search_agents_by_keyword(keyword, limit, offset)`	`List[AgentSearchResponse]`	Semantic keyword search
`search_agents_by_capabilities(capabilities, top_k)`	`List[AgentSearchResponse]`	Search by capability terms
`connect_agent(agent)`	`None`	Connect to another agent
`send_message(content, message_type, receiver_id)`	`str`	Send message to connected agent
`add_message_handler(handler_fn)`	`None`	Register incoming message handler
`set_response(message_id, response)`	`None`	Set sync response for a message
`set_agent_executor(executor)`	`None`	Set LangChain/LangGraph executor
`update_agent_connection_info()`	`None`	Sync connection info with registry
`verify_agent_identity(credential_document)`	`bool`	Verify agent DID credential

AgentMessage

python

AgentMessage(
    content: str,
    sender_id: str,
    sender_did: dict = None,
    receiver_id: str = None,
    message_type: str = "query",
    message_id: str = None,         # Auto-generated UUID
    conversation_id: str = None,    # Auto-generated UUID
    in_reply_to: str = None,
    metadata: dict = None
)

Method	Returns	Description
`to_dict()`	`dict`	Convert to dictionary (includes `content` and `prompt` fields)
`to_json()`	`str`	Convert to JSON string
`from_dict(data)`	`AgentMessage`	Class method: create from dict
`from_json(json_str)`	`AgentMessage`	Class method: create from JSON

X402PaymentProcessor

python

X402PaymentProcessor(agent_seed: str, max_payment_usd: float = 0.1)

Method	Returns	Description
`get(url, **kwargs)`	`requests.Response`	GET with auto-payment
`post(url, data, json, **kwargs)`	`requests.Response`	POST with auto-payment
`request(method, url, **kwargs)`	`requests.Response`	Any HTTP method with auto-payment
`close()`	`None`	Close session
`account.address`	`str`	Wallet address

Supports context manager (with statement).

SearchAndDiscoveryManager

Method	Returns	Description
`search_agents(keyword, name, capabilities, status, did, limit, offset)`	`List[AgentSearchResponse]`	Full search with all filters
`search_agents_by_keyword(keyword, limit, offset)`	`List[AgentSearchResponse]`	Simple keyword search
`search_agents_by_capabilities(capabilities, top_k)`	`List[AgentSearchResponse]`	Search by capability list
`get_agent_by_id(agent_id)`	`Optional[AgentSearchResponse]`	Get specific agent

AgentSearchResponse

python

class AgentSearchResponse(TypedDict):
    id: str
    name: str
    description: str
    mqttUri: Optional[str]
    httpWebhookUrl: Optional[str]
    inboxTopic: Optional[str]
    capabilities: Optional[dict]
    status: Optional[str]
    didIdentifier: str
    did: str  # JSON string of DID credential

ConfigManager

Method	Returns	Description
`load_config(config_dir)`	`dict or None`	Load saved config
`save_config(config, config_dir)`	`None`	Save config to file
`create_agent(registry_url, api_key, name, description, capabilities, config_dir)`	`dict`	Provision new agent via API
`load_or_create(agent_config)`	`dict`	Load existing or create new

IdentityManager

Method	Returns	Description
`verify_agent_identity(credential_document)`	`bool`	Verify DID credential via registry
`get_identity_document()`	`str`	Get identity document
`get_my_did()`	`dict`	Get agent's DID
`load_did(cred_path)`	`None`	Load DID from file

Encryption Utilities

zyndai_agent.utils

Function	Description
`encrypt_message(message, recipient_did)`	ECIES encryption using recipient's DID public key
`decrypt_message(encrypted_msg, seed, did)`	Decrypt with seed + DID validation
`derive_private_key_from_seed(seed)`	SHA-256 hash of base64-decoded seed
`extract_public_key_from_did(did_doc)`	Extract secp256k1 public key from AuthBJJ DID
`private_key_from_base64(seed_b64)`	Convert base64 seed to 0x-prefixed hex private key

API Reference ​

AgentConfig ​

ZyndAIAgent ​

Properties ​

Methods ​

AgentMessage ​

X402PaymentProcessor ​

SearchAndDiscoveryManager ​

AgentSearchResponse ​

ConfigManager ​

IdentityManager ​

Encryption Utilities ​

API Reference

AgentConfig

ZyndAIAgent

Properties

Methods

AgentMessage

X402PaymentProcessor

SearchAndDiscoveryManager

AgentSearchResponse

ConfigManager

IdentityManager

Encryption Utilities