contributing: Add more human responsibility in AI use#7120
Open
wenzeslaus wants to merge 3 commits intoOSGeo:mainfrom
Open
contributing: Add more human responsibility in AI use#7120wenzeslaus wants to merge 3 commits intoOSGeo:mainfrom
wenzeslaus wants to merge 3 commits intoOSGeo:mainfrom
Conversation
This modifies the agent and contributing instructions to back the code in the change and also pre-review the code before submitting it. While the contributing file already has instructions for taking responsibility, this add explicit ask for reviewing your own code before submitting. For the agent instructions, this pushes the agent more to do a review by including explicit instructions. The policy section and these instructions are moved to the top to increase the chances that the agent will pick them up and consider them, and are also formulated more as instructions in imperative (rather than policy). I'm doing the move because Claude was not picking up the instructions anymore even when asked after the file became longer. The no-AI-co-author instructions is now more connected with the acknowledge instruction hopefully providing more clarity to the intention.
petrasovaa
reviewed
Feb 25, 2026
echoix
reviewed
Feb 26, 2026
AGENTS.md
Outdated
Comment on lines
+10
to
+13
| comments, and commit messages should anticipate these questions — not by | ||
| adding explanatory fluff, but by making the code itself clear and | ||
| well-reasoned. If a line cannot be justified, change or remove it rather | ||
| than inventing a rationale for keeping it. |
Member
There was a problem hiding this comment.
I've read that we should avoid negative examples, and direct instructions are better respected. I assume it can mix up what to do. Here, there's a couple of "negative" that might be better to be direct? To try, as I don't really use these kinds of tools yet. I only myself did some kind of traditional machine learning/deep learning just before LLMs exploded, so I see the (mathematical) reasoning in terms of RNN and attention.
…nsibility-in-ai-usage-and-policy
echoix
approved these changes
Feb 26, 2026
Member
|
@wenzeslaus is it ready? |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This modifies the agent and contributing instructions to back the code in the change and also pre-review the code before submitting it. While the contributing file already has instructions for taking responsibility, this add explicit ask for reviewing your own code before submitting.
For the agent instructions, this pushes the agent more to do a review by including explicit instructions. The policy section and these instructions are moved to the top to increase the chances that the agent will pick them up and consider them, and are also formulated more as instructions in imperative (rather than policy). I'm doing the move because Claude was not picking up the instructions anymore even when asked after the file became longer.
The no-AI-co-author instructions is now more connected with the acknowledge instruction hopefully providing more clarity to the intention.