Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature request]: Make Patterns more discoverable #281

Open
askpatrickw opened this issue Mar 21, 2024 · 2 comments
Open

[Feature request]: Make Patterns more discoverable #281

askpatrickw opened this issue Mar 21, 2024 · 2 comments
Labels
enhancement New feature or request

Comments

@askpatrickw
Copy link

askpatrickw commented Mar 21, 2024

What do you need?

--list shows you the names of all the patterns, but other than roaming through the file system there is no way to discover patterns and what they do.

I think an output like this would be great.

pattern description
pattern_name A brief description of what this pattern does.

I'm not sure if the description can be a section in the system.md or if it needs to be separate.
Hmm... maybe summarize_pattern is a new pattern and --listpatterns just loops through all the pattern folders.

This output is a bit too verbose.
cat patterns/rate_content/system.md | fabric -s -p summarize

ONE SENTENCE SUMMARY:

This task outlines a method for classifying and rating content based on thematic relevance and idea quantity.

MAIN POINTS:

  1. Content is labeled with up to 20 single-word descriptors.
  2. Labels must be unique and relevant to the content.
  3. Content rating is based on idea quantity and thematic alignment.
  4. Themes include human meaning, AI's future, and abstract thinking.
  5. Ratings range from S Tier (highest) to D Tier (lowest).
  6. S Tier requires 18+ ideas or strong theme matching.
  7. A Tier is for 15+ ideas or good theme matching.
  8. B Tier applies to 12+ ideas or decent theme matching.
  9. C Tier is given for 10+ ideas or some theme matching.
  10. D Tier indicates few ideas or little theme relevance.

TAKEAWAYS:

  1. The process emphasizes both the quantity and quality of ideas.
  2. Thematic relevance is crucial for higher ratings.
  3. The system allows for nuanced content evaluation.
  4. Labels help categorize content for easier understanding.
  5. This method encourages in-depth analysis of content's value and relevance.%
@askpatrickw askpatrickw added the enhancement New feature or request label Mar 21, 2024
@askpatrickw
Copy link
Author

Updated the descriptions, I just discovered --list but I still think it could be improved.

@nopslip
Copy link

nopslip commented Mar 21, 2024

I've been thinking about this, too, and started playing with the idea of a 'meta-pattern' that creates categories and organizes the patterns into them. Here's a gist of what I was working on: https://gist.github.com/nopslip/e25b8d9f1d6e3dbbf0b61f2a0820890e

Of course, asking AI to create the categories using a 'meta-pattern' during the build pipeline would be prone to unknown results, so it's probably not ideal.

I like your idea here:

cat patterns/rate_content/system.md | fabric -s -p summarize

I also thought a simple web UI connected to the patterns folder in the repo could be cool. It could provide basic search and display of the readme's. For good form, maybe an example or two of the pattern in action, too.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants