johnzhang777/SpicyDiff πΆοΈ
LLM-powered code review with personality β ROAST or PRAISE your PRs! Supports OpenAI, DeepSeek, and more.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| github-token | GitHub token with pull-request write permission | yes | β |
| api-key | API key for the LLM provider | yes | β |
| provider | LLM provider shortcut: openai, deepseek, qwen, doubao, zhipu, moonshot, gemini, yi, baichuan, minimax | no | "" |
| model | LLM model name (optional β each provider has a sensible default) | no | "" |
| mode | Review personality β ROAST (brutally honest), PRAISE (blindly optimistic), or SECURITY (paranoid auditor) | no | ROAST |
| language | Output language: zh (δΈζ) or en (English) | no | en |
| base-url | Custom base URL for any OpenAI-compatible API (overrides provider preset) | no | "" |
| temperature | LLM sampling temperature (0.0 = deterministic, 1.0 = creative). Default: 0.7 | no | 0.7 |
| max-tokens | Maximum tokens in the LLM response. Default: 4096 | no | 4096 |
| max-diff-chars | Maximum diff size (chars) sent to the LLM. Larger diffs are truncated. Default: 60000 | no | 60000 |
| exclude-patterns | Comma-separated glob patterns to exclude from review (e.g. '*.test.js,docs/**') | no | "" |
| pr-number | Pull Request number (auto-detected from event context; set manually for workflow_dispatch) | no | "" |
| custom-rules | Comma-separated custom review rules to enforce (e.g. 'No hardcoded URLs,All functions must have docstrings') | no | "" |
| dry-run | If true, prints the review to logs without posting to GitHub. Default: false | no | false |
Outputs
no outputs