Run iFlow CLI commands within your GitHub workflows
github.com/iflow-ai/iflow-cli-action
A Docker-based action with Node.js 22, npm, and uv pre-installed
Pre-installed Node.js 22, npm, and uv for optimal performance
Support for iFlow API authentication with custom endpoints
Timeout control and working directory configuration
Summary reports in PRs and seamless issue/PR workflows
Add to your GitHub workflow:
- name: Run iFlow CLI
uses: iflow-ai/iflow-cli-action@v1.3.0
with:
prompt: "Analyze this codebase"
api_key: \${{ secrets.IFLOW_API_KEY }}
See documentation for detailed setup instructions
- name: iFlow CLI Issue Triage
uses: iflow-ai/iflow-cli-action@v1.3.0
with:
api_key: \${{ secrets.IFLOW_API_KEY }}
prompt: |
## Role
You are an issue triage assistant...
## Steps
1. Run: \`gh label list\` to get all available labels
2. Review the issue title and body
3. Classify by kind and priority
4. Apply labels with \`gh issue edit\`
5. Remove "status/needs-triage" if present
Provide complete settings.json
for advanced configuration
Connect to external tools via Model Context Protocol
Run setup commands before iFlow execution with precmd
Pass additional CLI arguments with extra_args
prompt: "Review this PR for quality, security, and best practices"
prompt: "/init && Generate comprehensive API documentation"
prompt: "Suggest a comprehensive testing strategy"
prompt: "Analyze system architecture for improvements"
Input | Description | Required |
---|---|---|
prompt |
The prompt to execute with iFlow CLI | Yes |
api_key |
iFlow API key for authentication | Yes |
base_url |
Custom base URL for iFlow API | No |
model |
Model name to use (default: Qwen3-Coder) | No |
timeout |
Timeout in seconds (1-86400) | No |
Get your API key at iflow.cn
Explore the possibilities with iFlow CLI GitHub Action
Documentation: iflow-ai.github.io/iflow-cli-action
Contribute: GitHub Repository
License: MIT