finegrained activities config customization

[cocktailpeanut]

Description

Was looking into why Twitter posts all sound the same, and the causes were:

1. Using 10 recent tweets (hardcoded): For example, if set to 1, should only reference 1 recent tweet.
2. Using hardcoded system prompts: For example instead of Please craft a short tweet (under 280 chars) that references these memories, reflects the personality and objectives, and ensures it's not repetitive or dull. Keep it interesting, cohesive, and mindful of the overall tone.', can do Please craft a short tweet (under 140 chars) that references these memories, reflects the personality and objectives, and ensures it's not repetitive or dull. Keep it interesting, cohesive, and mindful of the overall tone. No emojis. Do NOT ask questions, just make statements.`

Instead of hardcoding, I looked into customizing these attributes via the config, at the moment it looks like the activity_constraints.json is the best place for it, but let me know if there are better places to configure this.

Type of Change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• [O] New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation Update

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration.

1. Change the system_prompt or num_activities_to_fetch attribute from https://github.com/peanutcocktail/pippin/blob/main/my_digital_being/config_sample/activity_constraints.json#L68-L72 or https://github.com/peanutcocktail/pippin/blob/main/my_digital_being/config_sample/activity_constraints.json#L73-L77
2. Restart pippin.
3. It should now use that template instead of the previously hardcoded template.

Checklist:

• [O] My code follows the style guidelines of this project
• [O] I have performed a self-review of my own code
• [O] I have commented my code, particularly in hard-to-understand areas
• [O] I have made corresponding changes to the documentation
• [O] My changes generate no new warnings
• [O] Any dependent changes have been merged and published in downstream modules