ecodrix/Erix Buddy AI Code Reviewer
Bring-your-own-key AI code reviewer supporting Gemini, OpenAI, Anthropic, Ollama, Groq, and more.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Jun 26, 2026
- License
- MIT
Pinned Snippet
uses: ecodrix/erix_buddy@90024453ab1020121e77c83b97e77f905203e616 # code_reviewertags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| provider | LLM provider id: gemini | vertex | openai | anthropic | ollama | openrouter | groq | azure-openai | deepseek | mistral | xai | ecodrix | no | gemini |
| api-key | API key for the chosen provider. Use a repository secret. | no | — |
| model | Model id (provider-specific). Falls back to the provider's recommended default. | no | — |
| base-url | Override the provider's API base URL (Azure / Ollama / custom relays) | no | — |
| vertex-project | Google Cloud project id (only for provider=vertex) | no | — |
| vertex-location | Vertex AI region (only for provider=vertex) | no | us-central1 |
| azure-deployment | Azure OpenAI deployment name (only for provider=azure-openai) | no | — |
| mode | Review mode: branch | staged | last | unstaged | full | audit | no | branch |
| branch | Base branch ref to diff against in branch mode | no | origin/main |
| full-root | Subdirectory root for full-codebase audit mode | no | . |
| paths | Comma- or newline-separated path filters (regex/substring) for audit mode | no | — |
| max-files | Cap files reviewed in audit mode | no | 500 |
| severity | Minimum severity to report: critical | major | minor | nitpick | info | no | minor |
| categories | Comma-separated list of categories to surface (security,correctness,performance,maintainability,style,testing,documentation,architecture,accessibility,design-system). Defaults to all. | no | — |
| concurrency | Parallel file reviews | no | 4 |
| report-path | Markdown report output path | no | code-review-report.md |
| fail-on-blocking | Exit non-zero when critical/major issues are found | no | true |
| no-walkthrough | Skip the PR walkthrough generation phase (saves one LLM call) | no | false |
| no-mermaid | Skip the mermaid sequence diagram in the walkthrough | no | false |
| no-architecture | Skip the cross-cutting architecture pass (saves one LLM call) | no | false |
| no-tests | Skip generating test-coverage gap suggestions | no | false |
| custom-instructions | Free-form project-specific guidelines injected into the system prompt | no | — |
| custom-instructions-file | Path to a file in the repo whose contents become the customInstructions. Useful for multi-line prompts that don't fit cleanly in YAML. | no | — |
| post-pr-comment | Post (or update) a sticky PR comment with the markdown report | no | true |
| github-token | Override GITHUB_TOKEN for PR comment posting (e.g. when using a GitHub App token with elevated permissions) | no | — |
| verbose | Enable verbose logging in the underlying CLI (--verbose). Useful for debugging schema-validation retries. | no | false |
Outputs
| name | description |
|---|---|
| report-path | Path to the generated markdown report |