From 9ab7f1804a8f8211dd87b474d512203a559f0072 Mon Sep 17 00:00:00 2001 From: Jonathan Peppers Date: Fri, 6 Dec 2024 16:52:03 -0600 Subject: [PATCH] [actions] update to inclusive-heat-sensor v0.1.2 (#26443) Context: https://github.com/jonathanpeppers/inclusive-heat-sensor/releases/tag/v0.1.2 Changes: https://github.com/jonathanpeppers/inclusive-heat-sensor/compare/v0.1.1...v0.1.2 After reviewing comments coming in for the past month, we made a few adjustments to make things more precise. We also stopped recording comment text, and simply a use public link to it instead. v0.1.2 has these changes. Let's also configure the bot to auto-hide comments if they score a 9 out of 10 or higher. This seems reasonable as: * A comment like "This sucks" is a 7/10. So, it would have to be pretty bad (mean comment!) to get auto-hidden. * If the comment triggers a Microsoft/OpenAI "content filter", we ignore the comment entirely as we can't be sure it's offensive at all. We will continue monitoring the results. --- .github/workflows/inclusive-heat-sensor.yml | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/.github/workflows/inclusive-heat-sensor.yml b/.github/workflows/inclusive-heat-sensor.yml index 0fc0462aa3d3..26664076b285 100644 --- a/.github/workflows/inclusive-heat-sensor.yml +++ b/.github/workflows/inclusive-heat-sensor.yml @@ -14,5 +14,9 @@ permissions: jobs: detect-heat: - uses: jonathanpeppers/inclusive-heat-sensor/.github/workflows/comments.yml@v0.1.1 + uses: jonathanpeppers/inclusive-heat-sensor/.github/workflows/comments.yml@v0.1.2 secrets: inherit + with: + minimizeComment: true + offensiveThreshold: 9 + angerThreshold: 9