Gregor Martynus

if you put it at the end it could override the options we pass in like `genReqId`. But maybe we can move the `logger` option above `...options`?

> I can't find any code doing any cache handling on the Octokit side The only cache behavior we do is to throw an error in order to implement caching:...

@kfcampbell @wolfy1339 the ergonomics of GHES / GHEC endpoints are much better implemented in https://github.com/octokit/octokit.js, see this update https://github.com/octokit/octokit.js/issues/2127#issuecomment-922150660. It's hard to make types work well for targets other than...

I agree that using `git push --dry-run [REPO_URL] --tags` to check for write access makes more sense to me than what we doing today.

> There is already a test for this case Ha, what do you know 🤣 I think we should 1. Implement this fix in https://github.com/gr2m/universal-github-app-jwt 2. Remove the special error...

Hmm sorry some of the tests are failing, from the looks of it, it is most likely related to #857. Could you have a look? I'm sorry this is probably...

> requires some new dependencies you don't need the `@octokit/graphql` dependency to mock graphql requests, you just mock the `POST /graphql` request like any other.

Holler if you need help with this PR

I ran it locally and got the following error ``` [8:46:26 PM] [semantic-release] › ℹ Start step "success" of plugin "@semantic-release/github" [8:46:27 PM] [semantic-release] › ✘ Failed step "success" of...

It's possible that this error is an edge case that only happens because I ran it locally. The commit only exist on my machine due to the local testing. When...