Skip to content

Conversation

kotamat
Copy link

@kotamat kotamat commented Jun 26, 2025

Awesome Claude Code Contribution

Thank you for taking time to submit to awesome-claude-code! We appreciate any submission, whether you are the author or not. In the current state, @hesreallyhim is the sole maintainer, and if there are any issues with your submission, these can be discussed in the PR comments.

Type of Contribution

<!-- Mark the type of contribution with an [x] -->

  • Slash-command example
  • CLAUDE.md file example
  • Workflow example
  • awesome-claude-code repository improvement/fix
  • Other/hard-to-classify.

Source Information

Summary

Adds a Python script (fetch_commands.py) that automatically extracts and downloads all slash commands from the awesome-claude-code README to the user's ~/.claude/commands/ directory. This makes it easy for users to install all available commands with a single script execution, improving the user experience for adopting the curated slash commands.

Verification Checklist

Proposing a change to the Awesome Claude Code repository itself

  • Is a PR to improve some aspect of the repository itself, and is not a contribution of a resource to the list.

Additional Context

The script handles various GitHub URL formats (blob URLs, gists) and converts them to raw content URLs for downloading. It includes error handling and rate limiting to be respectful of GitHub's servers.

🤖 Generated with Claude Code

kotamat and others added 2 commits June 26, 2025 22:21
This script extracts all slash command URLs from the README.md file and automatically downloads them to the user's ~/.claude/commands directory, making it easy to install all available commands at once.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <[email protected]>
Added entry to Tooling section documenting the Python script that automatically downloads slash commands from the awesome list to local ~/.claude/commands directory.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <[email protected]>
Copy link
Owner

@hesreallyhim hesreallyhim left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@kotamat awesome contribution, thanks for taking the time to do this. i will run it asap and get back to you with any comments, but this is very cool. I'm wondering - do you think it would be valuable to do the same thing for CLAUDE.md files? this also sparks an idea of turning this awesome-list into an MCP that could provide users with specific commands for specific cases, or help optimize a command/etc... so i'll try to get to this very soon, thanks for the effort.

In the meantime, can you just add a brief HOW-TO at the top? It looks like this is configured to run off a cloned copy of this repo, but I'd like to be sure what you intend. And also it might be cool to add an option to just pull down the README, keep it in memory while processing, and then release it. But yeah just some really basic setup directions would be helpful.

@kotamat
Copy link
Author

kotamat commented Jun 27, 2025

@hesreallyhim
Thanks for the positive feedback! I'm excited about the MCP idea too - that would be a natural evolution of this concept.

You're absolutely right about needing clearer setup instructions. Yes, the script is currently configured to run from a cloned copy of this repo since it
reads the local README.md file to extract the slash command URLs.

I'd be happy to add a brief HOW-TO section at the top of the README. For the memory-based approach you mentioned, that's a great suggestion - it would make
the tool more portable by fetching the README directly from GitHub rather than requiring a local clone.

Regarding CLAUDE.md files, that would definitely be valuable! The same parsing approach could work well for extracting and organizing those resources too.

Let me know if you'd like me to implement any of these improvements - I can add the HOW-TO section and potentially the remote README fetching option in a
follow-up commit.

- Add comprehensive Quick Start section to README with step-by-step instructions
- Support both local and remote execution methods
- Add --remote flag to fetch README.md directly from GitHub
- Update to use uv for automatic dependency resolution
- Include alternative installation method without cloning
- Remove redundant fetch_commands.py entry from Tooling section

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <[email protected]>
@kotamat kotamat requested a review from hesreallyhim June 27, 2025 09:23
@hesreallyhim hesreallyhim self-requested a review June 28, 2025 15:36
Copy link
Owner

@hesreallyhim hesreallyhim left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

hey you might want to take another look at the design because i've added a csv file with metadata for all the resources, prob removes the need or at least changes the logic for parsing the README vs. just reading off the CSV. also, fwiw, i'm probably going to just host copies of all of the files that the licenses permit me to. One advantage of your script though is you can download all the resources even if I can't host them on this repo.

https://github.com/hesreallyhim/awesome-claude-code/blob/main/.myob/scripts/resource-metadata.csv

@hesreallyhim
Copy link
Owner

@kotamat could you take a look at the new repo structure and let me know if you think there is still value in this PR? I really appreciate the effort, but I decided to host the resources for posterity (at least the ones that I'm allowed to). I think one advantage of your script is that it would allow people to download resources that I can't host, so let me know what you think - or if you want to work on an MCP or something, let me know.

@hesreallyhim hesreallyhim marked this pull request as draft July 11, 2025 03:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants