AI Modules should not default to or **require** AgentAPI/Tasks

[bpmct]

There is a lot of value in using AI agents like Claude Code without Tasks/AgentAPI, including pre-configuring it the CLI.

However, I believe our modules are designed primarily with the intention of working with tasks and it can be difficult to use it standalone.

Proposal

• For top 3 modules, clearly delineate how to use it standalone vs. with tasks in the docs

Some context

[matifali]

For the test, make sure we pass the following use cases

1. When using with Tasks (agentAPI), if a task workspace is restarted, the task should continue from the previous session.
2. When using standalone (no agnetAPI), only start if a user explicitly starts the agent (no start on workspace startup)
3. When using standalone (no agnetAPI), try to resume the last session if it exists.