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鈥檒l occasionally send you account related emails.

Already on GitHub? Sign in to your account

Build the framework to be able to create a failing test simply from the logs/DEBUG people report #3847

Closed
1 task done
waynehamadi opened this issue May 5, 2023 · 6 comments

Comments

@waynehamadi
Copy link
Contributor

waynehamadi commented May 5, 2023

Duplicates

  • I have searched the existing issues

Summary 馃挕

In order to improve Auto-GPT we have to fix user's bugs quicker and more efficiently.

To do that, we want to create a battery of failing tests that showcase some of Auto-GPT's inconsistent behaviors or bugs.

We want to create tests simply from the issues people report:

  • step 1 the user reports an issue and puts the logs/DEBUG folder as a zip
  • we download the ZIP, put it in our testing suite and "replay" the bug.
  • we create a pull request and once it gets merged, the community can fix the bug.

This issue requires good understanding of pytest

DM me on discord : merwanehamadi

Join Auto-GPT's channel so we can talk about Auto-GPT: https://discord.gg/autogpt

Examples 馃寛

No response

Motivation 馃敠

No response

@anonhostpi
Copy link

Man your competence in Software QA is through the roof.

This probably seems like a basic question to you, but the design principle behind these tests is to try to test any bug fixes against the situation that discovered them?

@anonhostpi
Copy link

If that is the case, this framework might be helpful for towards y'alls efforts of making AutoGPT capable of self-improvement -link to my gists tracking y'alls issues/PRs/discussions

@Boostrix
Copy link
Contributor

Boostrix commented May 7, 2023

we want to create a battery of failing tests that showcase some of Auto-GPT's inconsistent behaviors or bugs.

To get this started, one of the lowest-hanging fruits is probably the write_to_file issue where dupe-checking sees in the log file that the file was previously written by another agent instance, and fails to write/update the file because of that - despite the file possibly no longer existing (or despite it being outdated).

Details: #3514

@waynehamadi
Copy link
Contributor Author

@anonhostpi that's it ! this is the place I want to to go to:

  • User finds an issue
  • User uploads his replay
  • we create the "Fix Auto-GPT" challenge for it
  • people solve it
  • we know it won't appear again because a test protects it.

@github-actions
Copy link
Contributor

github-actions bot commented Sep 6, 2023

This issue has automatically been marked as stale because it has not had any activity in the last 50 days. You can unstale it by commenting or removing the label. Otherwise, this issue will be closed in 10 days.

@github-actions github-actions bot added the Stale label Sep 6, 2023
@github-actions
Copy link
Contributor

This issue was closed automatically because it has been stale for 10 days with no activity.

@github-actions github-actions bot closed this as not planned Won't fix, can't repro, duplicate, stale Sep 17, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants