agentconnect.config.models module¶

Pydantic configuration models for AgentConnect.

This module defines the comprehensive configuration structure using Pydantic models. Only essential, developer-facing settings are exposed - internal tuning parameters remain in the code as defaults.

Global configuration precedence (highest to lowest): 1) Runtime keyword overrides 2) agentconnect.yaml 3) Model defaults

Environment variables are not read by the global settings system. They are reserved for secrets and are consumed directly by the specific subsystems that require them (e.g., provider SDKs, external clients).

class InMemoryConfig(**data)¶

Bases: BaseModel

In-memory Qdrant configuration (for development).

Parameters:: type (Literal['in_memory'])

type: Literal['in_memory']¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class LocalFileConfig(**data)¶

Bases: BaseModel

Local file-based Qdrant configuration.

Parameters:

type (Literal['local_file'])
path (str)

type: Literal['local_file']¶

path: str¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RemoteConfig(**data)¶

Bases: BaseModel

Remote Qdrant server configuration.

Parameters:

type (Literal['remote'])
url (str)
api_key (SecretStr)

type: Literal['remote']¶

url: str¶

api_key: SecretStr¶

classmethod validate_url_format(v)¶

Validate URL format.

Return type:: str
Parameters:: v (str)

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class VectorSearchAdvancedSettings(**data)¶

Bases: BaseModel

Advanced configuration for vector search (for power users).

Parameters:

timeout (int)
grpc_port (int | None)
prefer_grpc (bool)
use_quantization (bool)
vectors_on_disk (bool)
index_on_disk (bool)
batch_size (int)

timeout: int¶

grpc_port: Optional[int]¶

prefer_grpc: bool¶

use_quantization: bool¶

vectors_on_disk: bool¶

index_on_disk: bool¶

batch_size: int¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class VectorSearchSettings(**data)¶

Bases: BaseModel

Vector search configuration for agent registry.

Parameters:

model_name (str)
cache_folder (str)
vector_store_path (str)
deployment (InMemoryConfig | LocalFileConfig | RemoteConfig)
advanced (VectorSearchAdvancedSettings)

model_name: str¶

cache_folder: str¶

vector_store_path: str¶

deployment: Union[InMemoryConfig, LocalFileConfig, RemoteConfig]¶

advanced: VectorSearchAdvancedSettings¶

get_connection_config()¶

Get connection configuration for the vector implementation.

Return type:: Dict[str, Any]

get_performance_config()¶

Get performance-related configuration.

Return type:: Dict[str, Any]

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RegistrySettings(**data)¶

Bases: BaseModel

Registry subsystem configuration.

The registry only needs vector-search / storage related configuration.

Runtime parameters for the standalone Registry API server are configured separately in the agentconnect.servers package via environment variables with the AGENTCONNECT_REGISTRY_ prefix. See agentconnect.servers.config.RegistryAPISettings.

Parameters:: vector_search (VectorSearchSettings)

vector_search: VectorSearchSettings¶

get_vector_search_config()¶

Get vector search configuration as dictionary for backward compatibility.

Return type:: Dict[str, Any]

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class CommunicationSettings(**data)¶

Bases: BaseModel

Communication hub configuration for A2A messaging.

Parameters:: enable_message_history (bool)

enable_message_history: bool¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RegistryClientSettings(**data)¶

Bases: BaseModel

Configuration for the Registry API Client.

Parameters:

base_url (str)
default_timeout (float)
connect_timeout (float)
read_timeout (float)
pool_timeout (float)
max_retries (int)
retry_backoff_factor (float)
retryable_status_codes (List[int])
max_connections (int)
max_keepalive_connections (int)

base_url: str¶

default_timeout: float¶

connect_timeout: float¶

read_timeout: float¶

pool_timeout: float¶

max_retries: int¶

retry_backoff_factor: float¶

retryable_status_codes: List[int]¶

max_connections: int¶

max_keepalive_connections: int¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class ClientSettings(**data)¶

Bases: BaseModel

Configurations for API clients.

Parameters:: registry (RegistryClientSettings)

registry: RegistryClientSettings¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class MCPAgentDiscoverySettings(**data)¶

Bases: BaseModel

Configuration for the Agent Discovery MCP tool.

Parameters:

enabled (bool)
top_k (int)
strictness (float)
output_detail (Literal['minimal', 'summary', 'capabilities', 'full'])

enabled: bool¶

top_k: int¶

strictness: float¶

output_detail: Literal['minimal', 'summary', 'capabilities', 'full']¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class MCPSettings(**data)¶

Bases: BaseModel

Configuration for MCP (Model Context Protocol) servers.

Parameters:: agent_discovery (MCPAgentDiscoverySettings)

agent_discovery: MCPAgentDiscoverySettings¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class PaymentsSettings(**data)¶

Bases: BaseModel

Payments configuration.

Parameters:

default_token_symbol (str)
wallet_data_dir (str)

default_token_symbol: str¶

wallet_data_dir: str¶

classmethod normalize_token_symbol(v)¶

Normalize token symbols to uppercase for consistent comparisons

Return type:: str
Parameters:: v (str)

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class LoggingSettings(**data)¶

Bases: BaseModel

Logging configuration settings.

Not part of global settings; retained for potential reuse.

Parameters:

level (str)
format (str)
json_output (bool)

level: str¶

format: str¶

json_output: bool¶

classmethod validate_log_level(v)¶

Validate log level is supported.

Return type:: str
Parameters:: v (str)

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class AgentConnectSettings(**data)¶

Bases: BaseModel

Main configuration class for AgentConnect.

Precedence (highest to lowest): 1. Runtime kwargs 2. agentconnect.yaml file 3. Hard-coded defaults

Parameters:

registry (RegistrySettings)
communication (CommunicationSettings)
clients (ClientSettings)
mcp (MCPSettings)
payments (PaymentsSettings)
project_name (str)

registry: RegistrySettings¶

communication: CommunicationSettings¶

clients: ClientSettings¶

mcp: MCPSettings¶

payments: PaymentsSettings¶

project_name: str¶

classmethod create_from_dict(config_dict)¶

Create settings instance from configuration dictionary.

Return type:: AgentConnectSettings
Parameters:: config_dict (Dict[str, Any])

get_registry_config()¶

Get registry configuration for backward compatibility.

Return type:: Dict[str, Any]

model_dump_yaml_safe()¶

Export configuration in YAML-safe format (no secrets).

Return type:: Dict[str, Any]

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].