Fix: Module init and agents

[leshy]

No description provided.

[chatgpt-codex-connector]

Codex usage limits have been reached for code reviews. Please check with the admins of this repo to increase the limits by adding credits.
Credits must be used to enable repository wide code reviews.

[greptile-apps]

Greptile Summary

This PR fixes the module initialization in SkillCoordinator by properly implementing inheritance from ModuleBase. The changes make SkillCoordinatorConfig extend ModuleConfig and update the __init__ method to call super().__init__(**kwargs) instead of manually managing event loops and skill container initialization. This resolves a TODO comment about proper use of super() and eliminates code duplication by relying on the parent class's initialization logic.

• Made SkillCoordinatorConfig inherit from ModuleConfig
• Updated SkillCoordinator.__init__() to properly call super().__init__(**kwargs)
• Removed duplicate _loop and _loop_thread instance variable declarations that are now inherited from ModuleBase
• Removed manual get_loop() call, delegating to parent class initialization

Confidence Score: 5/5

• This PR is safe to merge with no concerns. It properly refactors initialization to follow Python inheritance patterns.
• Score reflects that this is a straightforward refactoring that fixes improper initialization patterns. The changes correctly delegate to parent class constructors, which properly initialize _loop and _loop_thread through ModuleBase.__init__(). All existing usages in the codebase (agents2/agent.py, skillspy.py, test_coordinator.py) instantiate SkillCoordinator() with no arguments, which works correctly since **kwargs accepts zero arguments. The inheritance chain is well-tested through existing test suite.
• No files require special attention

Important Files Changed

Filename

Overview

dimos/protocol/skill/coordinator.py

Fixed module initialization by making SkillCoordinatorConfig inherit from ModuleConfig and updating init to properly call super().init(**kwargs) instead of manually calling SkillContainer.init() and get_loop(). Removed duplicate _loop/_loop_thread attributes that are now properly inherited from ModuleBase. Changes align with the module's parent class hierarchy and fix a TODO comment about proper inheritance.

Sequence Diagram

sequenceDiagram
    participant Client
    participant SkillCoordinator
    participant ModuleBase
    participant Configurable
    participant get_loop

    Client->>SkillCoordinator: __init__(**kwargs)
    SkillCoordinator->>ModuleBase: super().__init__(**kwargs)
    ModuleBase->>Configurable: super().__init__(**kwargs)
    Configurable->>Configurable: self.config = ModuleConfig(**kwargs)
    Configurable-->>ModuleBase: return
    ModuleBase->>get_loop: get_loop()
    get_loop-->>ModuleBase: (loop, thread)
    ModuleBase->>ModuleBase: self._loop = loop<br/>self._loop_thread = thread
    ModuleBase-->>SkillCoordinator: return
    SkillCoordinator->>SkillCoordinator: self._static_containers = []<br/>self._dynamic_containers = []<br/>self._skills = {}<br/>self._skill_state = SkillStateDict()
    SkillCoordinator-->>Client: initialized

Loading

[greptile-apps]

Greptile's behavior is changing!

From now on, if a review finishes with no comments, we will not post an additional "statistics" comment to confirm that our review found nothing to comment on. However, you can confirm that we reviewed your changes in the status check section.

_{This feature can be toggled off in your Code Review Settings by deselecting "Create a status check for each PR".}