dataflow.registry¶
Haive Registry System.
This package provides a comprehensive registry system for Haive components, enabling discovery, registration, and management of various components like agents, tools, engines, and games.
The registry system acts as a central repository for component metadata, configurations, and dependency information, making it easier to discover and use components in the Haive ecosystem.
- Key Modules:
core: Core registry system implementation and API models: Data models for registry entries, configurations, and dependencies discovery: Component discovery mechanisms serialization: Utilities for serializing and deserializing complex objects providers: Entity type-specific provider implementations importers: Data importers for external component sources
Typical usage example:
>>> from haive.dataflow.registry import registry_system, EntityType, discover_agents
>>>
>>> # Discover and register all agents
>>> discovered_agents = discover_agents()
>>> print(f"Discovered {len(discovered_agents)} agents")
>>>
>>> # Get all registered tools
>>> tools = registry_system.get_entities_by_type(EntityType.TOOL)
>>> for tool in tools:
>>> print(f"Tool: {tool.name} - {tool.description}")
Submodules¶
- dataflow.registry.base
- dataflow.registry.bin
- dataflow.registry.core
- dataflow.registry.db
- dataflow.registry.discovery
- dataflow.registry.importers
- dataflow.registry.lazy_core
- dataflow.registry.models
- dataflow.registry.providers
- dataflow.registry.registries
- dataflow.registry.serialization
- dataflow.registry.utils