stillrivercode/AI Workflow Automation
AI-powered GitHub workflow automation featuring PR review with native resolvable suggestions. Extensible platform for intelligent code analysis and automation.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| github_token | GitHub token for API requests. | yes | — |
| openrouter_api_key | OpenRouter API key. | yes | — |
| openrouter_url | The URL of the OpenRouter API. | no | https://openrouter.ai/api/v1/chat/completions |
| model | The AI model to use for the review. Common examples: - google/gemini-2.5-pro (default, balanced performance) - anthropic/claude-3.5-sonnet (balanced performance) - anthropic/claude-3-opus (highest quality, more expensive) - anthropic/claude-3-haiku (fastest, most cost-effective) - openai/gpt-4o (OpenAI's latest) - openai/gpt-4o-mini (cost-effective OpenAI option) - google/gemini-pro (Google's flagship model) For the complete list of available models, see: https://openrouter.ai/models | no | google/gemini-2.5-pro |
| review_type | The type of review to perform (e.g., full, security, performance). | no | full |
| max_tokens | The maximum number of tokens for the AI model to generate. Must be between 1 and 32768 (model-dependent). | no | 4096 |
| temperature | The temperature for the AI model. Controls randomness: 0.0 = deterministic, 2.0 = very random. Must be between 0.0 and 2.0. | no | 0.7 |
| request_timeout_seconds | Timeout for the OpenRouter API request in seconds. Must be between 1 and 600. | no | 120 |
| retries | The number of times to retry the OpenRouter API request on failure. Must be between 0 and 5. | no | 3 |
| custom_review_rules | Path to a JSON file containing custom review rules. Currently only JSON format is supported. | no | — |
| exclude_patterns | A comma-separated list of glob patterns for files to exclude from the review. | no | — |
| post_comment | Whether to automatically post the review as a PR comment. When false, only outputs are set. | no | true |
Outputs
| name | description |
|---|---|
| review_comment | The content of the AI-generated review comment. |
| review_status | The status of the review (e.g., success, failure). |