| model-api-base | Base URL for an OpenAI-compatible API (e.g. vLLM on OpenShift AI, Gemini, etc.) | yes | — |
| model-api-key | API key for the model endpoint (leave empty if not required) | no | "" |
| model-name | Model name to use for inference (e.g. meta-llama/Llama-3.1-8B-Instruct, gemini-2.0-flash) | yes | — |
| docs-repo-url | URL of the documentation repository | yes | — |
| github-token | GitHub token for creating PRs | yes | — |
| dry-run | Simulate changes without writing files or pushing PR | no | false |
| pr-number | Pull request number | no | — |
| pr-base | Base branch for PR comparison (e.g. origin/main) | no | origin/main |
| pr-head-sha | Head reference/SHA for the PR | no | — |
| docs-subfolder | Optional: relative path to docs subfolder within the same repository (e.g., "docs", "content/docs") | no | "" |
| docs-base-branch | Base branch for docs repository PRs (e.g., "main" or "master") | no | main |
| comment-body | The comment body text (used to detect [review-docs], [update-docs], or [review-feature] commands) | no | "" |
| jira-url | Jira instance URL (e.g., https://your-company.atlassian.net). Required for [review-feature]. | no | "" |
| jira-username | Jira username/email for API authentication. Required for [review-feature]. | no | "" |
| jira-api-token | Jira API token for authentication. Required for [review-feature]. | no | "" |
| google-sa-key | Optional: Google service account JSON key for fetching Google Docs linked in Jira tickets. If not provided, Google Docs links will be flagged but content will not be fetched. | no | "" |
| max-context-chars | Maximum characters for LLM prompt content (default: 400000, ~100K tokens). Decrease for models with smaller context windows (e.g. 32000 for an 8K-token model). | no | 400000 |
| style-config-path | Path to a Markdown style configuration file (.md) containing documentation style guidelines. If not set, auto-detects .code-to-docs/style.md in the repository root. | no | "" |