ChatGuard is a GitHub Action designed to leverage ChatGPT for sentiment analysis of comments and discussions within your repository. It enhances community management by identifying and handling content that includes spam, advertisements, or offensive language. ChatGPT provides valuable insights into why specific comments are flagged.
.github/workflows/chatguard.yml
).name: ChatGuard
on:
issues:
types:
- opened
issue_comment:
types:
- created
- edited
pull_request_review_comment:
types:
- created
- edited
Add the ChatGuard action to your workflow, specifying the necessary parameters.
jobs:
analyze_content:
runs-on: ubuntu-latest
steps:
- name: Analyze Comments with ChatGPT
uses: nikpivkin/chat-guard@v0
with:
gh-token: ${{ secrets.GITHUB_TOKEN }}
openai-api-key: ${{ secrets.OPENAI_API_KEY }}
Name | Description | Default |
---|---|---|
gh-token |
GitHub personal access token. | |
openai-api-key |
API token for ChatGPT. | Required |
gpt-model |
The ChatGPT model used. See available models here. | "gpt-3.5-turbo" |
timeout |
Timeout in seconds. | 60 |
This project is licensed under the MIT License.