Skip to content

Conversation

bouzaghrane
Copy link
Contributor

@bouzaghrane bouzaghrane commented Dec 24, 2024

Modified create_file_content_string() to show root level README.md first in the repository content output file using a two pass approach. This is motivated by the structure proposed by AnswerDotAI for llms.txt files.

Ran the test suite and all 11 tests pass.

@bouzaghrane bouzaghrane changed the title Place ingested repository readme first in file content output Place repository root level readme first in file content output Dec 24, 2024
@cyclotruc
Copy link
Member

Thank you very much and thanks for the discovery of llms.txt files
I'm merging this and I'll probably add later improvement around this idea

@cyclotruc cyclotruc merged commit b315d52 into coderamp-labs:main Dec 25, 2024
4 checks passed
@rawwerks
Copy link

rawwerks commented Feb 12, 2025

hey @cyclotruc - i think "llms.txt" would be a really easy addition / mode to add to gitingest: AnswerDotAI/llms-txt#37

the format is gaining in popularity.

for example, i could pay to crawl my own website: https://llmstxt.firecrawl.dev/

or i could just do something like gitingest --llmstxt on the code base to generate this directly.

@cyclotruc
Copy link
Member

@rawwerks that's a very good point, but do you mean add a llm.txt to gitingest.com or an option to generate llm.txt files with gitingest?

I'm not sure what I would need to include in the digest to make it usefull for this format, would you have any clue on that?

@rawwerks
Copy link

i mean "add an option to generate llm.txt files with gitingest"

here's the spec => https://github.com/AnswerDotAI/llms-txt

  • An H1 with the name of the project or site. This is the only required section
  • A blockquote with a short summary of the project, containing key information necessary for understanding the rest of the file
  • Zero or more markdown sections (e.g. paragraphs, lists, etc) of any type except headings, containing more detailed information about the project and how to interpret the provided files
  • Zero or more markdown sections delimited by H2 headers, containing "file lists" of URLs where further detail is available
  • Each "file list" is a markdown list, containing a required markdown hyperlink name, then optionally a : and notes about the file.

Here is a mock example:

# Title > Optional description goes here Optional details go here ## Section name - [Link title](https://link_url): Optional link details ## Optional - [Link title](https://link_url) 

Note that the "Optional" section has a special meaning---if it's included, the URLs provided there can be skipped if a shorter context is needed. Use it for secondary information which can often be skipped.

FOLKS-Tech pushed a commit to FOLKS-Tech/gitingest that referenced this pull request Sep 5, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
3 participants