[GH-ISSUE #2390] [COLLABORATORS] We should impose a time limit before closing a PR with changes requested #53176

New Issue

2026-07-11T19:49:58-05:00

GiteaMirror commented

2026-07-11 19:49:58 -05:00

Originally created by @DrKJeff16 on GitHub (Jun 30, 2026).
Original GitHub issue: https://github.com/rockerBOO/awesome-neovim/issues/2390

Originally assigned to: @rockerBOO, @Penaz91, @DrKJeff16 on GitHub.

Reasoning

Recently we've recieved an influx of questionable PRs:

An ad-hoc time limit was imposed for such plugins, which (expectedly) ended up in such PRs getting closed.

If willing to impose a limit, should it be two days? Three days? What's a reasonable time limit for you?

Originally created by @DrKJeff16 on GitHub (Jun 30, 2026). Original GitHub issue: https://github.com/rockerBOO/awesome-neovim/issues/2390 Originally assigned to: @rockerBOO, @Penaz91, @DrKJeff16 on GitHub. ## Reasoning Recently we've recieved an influx of questionable PRs: - #2385 - #2379 - #2371 An _ad-hoc_ time limit was imposed for such plugins, which (expectedly) ended up in such PRs getting closed. If willing to impose a limit, should it be two days? Three days? What's a reasonable time limit for you?

GiteaMirror added the discussion label 2026-07-11 19:49:58 -05:00

GiteaMirror commented

2026-07-11 19:50:00 -05:00

@Penaz91 commented on GitHub (Jun 30, 2026):

I'm partial to 48 hours, cause that's my "ad-hoc solution", but I'm open to ideas.

@Penaz91 commented on GitHub (Jun 30, 2026): I'm partial to 48 hours, cause that's my "ad-hoc solution", but I'm open to ideas.

GiteaMirror commented

2026-07-11 19:50:00 -05:00

@rockerBOO commented on GitHub (Jun 30, 2026):

So is the issue with up-front work to review these that don't follow the rules? That they may be using an agent to make the PR (and might never come back to review any changes)? What is the downside from closing the PR but allow them to re-open if needed?

I feel for humans it might be appropriate to wait a week as they may only have time on the weekend or something in a week to get things updated.

So I would say there is an AI Agent suspected workflow where we should highlight they must follow the template. The issue the agents run into is not getting the template they should be using before making a PR (they are not using the web interface). So maybe some statement like follow the prompt template at .github/pull_request_template.md and will need to auto-close after 2 days. And PR's could be re-opened at the maintainer's best effort if the rules are later followed.

And for regular PR's a week still feels appropriate, with an expiring nice message saying if they make the changes later we can re-open or they could re-add it the PR.

That's just from my point of view so open to what are the pain points and what we are trying to fix without it being too hard for busy people.

@rockerBOO commented on GitHub (Jun 30, 2026): So is the issue with up-front work to review these that don't follow the rules? That they may be using an agent to make the PR (and might never come back to review any changes)? What is the downside from closing the PR but allow them to re-open if needed? I feel for humans it might be appropriate to wait a week as they may only have time on the weekend or something in a week to get things updated. So I would say there is an AI Agent suspected workflow where we should highlight they must follow the template. The issue the agents run into is not getting the template they should be using before making a PR (they are not using the web interface). So maybe some statement like follow the prompt template at [.github/pull_request_template.md](https://github.com/rockerBOO/awesome-neovim/blob/main/.github/pull_request_template.md) and will need to auto-close after 2 days. And PR's could be re-opened at the maintainer's best effort if the rules are later followed. And for regular PR's a week still feels appropriate, with an expiring nice message saying if they make the changes later we can re-open or they could re-add it the PR. That's just from my point of view so open to what are the pain points and what we are trying to fix without it being too hard for busy people.

GiteaMirror commented

2026-07-11 19:50:01 -05:00

@DrKJeff16 commented on GitHub (Jun 30, 2026):

I agree with @rockerBOO. I'm not so accustomed to identifying agents myself, but at the same time they're not that hard to spot(?)

@DrKJeff16 commented on GitHub (Jun 30, 2026): I agree with @rockerBOO. I'm not so accustomed to identifying agents myself, but at the same time they're not that hard to spot(?)

GiteaMirror commented

2026-07-11 19:50:01 -05:00

@Penaz91 commented on GitHub (Jun 30, 2026):

One week sounds sensible. I don't think agents will suddenly start responding and fixing their own PRs if we give them a week instead of 2 days.

And even if they did, it would fix the problem anyway.

@Penaz91 commented on GitHub (Jun 30, 2026): One week sounds sensible. I don't think agents will suddenly start responding and fixing their own PRs if we give them a week instead of 2 days. And even if they did, it would fix the problem anyway.

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/awesome-neovim#53176